<p>Rclone is a command line program to manage files on cloud storage. It is a feature rich alternative to cloud vendors' web storage interfaces. <ahref="#providers">Over 40 cloud storage products</a> support rclone including S3 object stores, business & consumer file storage services, as well as standard transfer protocols.</p>
<p>Rclone has powerful cloud equivalents to the unix commands rsync, cp, mv, mount, ls, ncdu, tree, rm, and cat. Rclone's familiar syntax includes shell pipeline support, and <code>--dry-run</code> protection. It is used at the command line, in scripts or via its <ahref="/rc">API</a>.</p>
<p>Users call rclone <em>"The Swiss army knife of cloud storage"</em>, and <em>"Technology indistinguishable from magic"</em>.</p>
<p>Rclone really looks after your data. It preserves timestamps and verifies checksums at all times. Transfers over limited bandwidth; intermittent connections, or subject to quota can be restarted, from the last good file transferred. You can <ahref="https://rclone.org/commands/rclone_check/">check</a> the integrity of your files. Where possible, rclone employs server-side transfers to minimise local bandwidth use and transfers from one provider to another without using local disk.</p>
<p>Virtual backends wrap local and cloud file systems to apply <ahref="https://rclone.org/crypt/">encryption</a>, <ahref="https://rclone.org/cache/">caching</a>, <ahref="https://rclone.org/compress/">compression</a><ahref="https://rclone.org/chunker/">chunking</a> and <ahref="https://rclone.org/union/">joining</a>.</p>
<p>Rclone <ahref="https://rclone.org/commands/rclone_mount/">mounts</a> any local, cloud or virtual filesystem as a disk on Windows, macOS, linux and FreeBSD, and also serves these over <ahref="https://rclone.org/commands/rclone_serve_sftp/">SFTP</a>, <ahref="https://rclone.org/commands/rclone_serve_http/">HTTP</a>, <ahref="https://rclone.org/commands/rclone_serve_webdav/">WebDAV</a>, <ahref="https://rclone.org/commands/rclone_serve_ftp/">FTP</a> and <ahref="https://rclone.org/commands/rclone_serve_dlna/">DLNA</a>.</p>
<p>Rclone is mature, open source software originally inspired by rsync and written in <ahref="https://golang.org">Go</a>. The friendly support community are familiar with varied use cases. Official Ubuntu, Debian, Fedora, Brew and Chocolatey repos. include rclone. For the latest version <ahref="https://rclone.org/downloads/">downloading from rclone.org</a> is recommended.</p>
<p>Rclone is widely used on Linux, Windows and Mac. Third party developers create innovative backup, restore, GUI and business process solutions using the rclone command line or API.</p>
<p>Rclone does the heavy lifting of communicating with cloud storage.</p>
<li>Backup (and encrypt) files to cloud storage</li>
<li>Restore (and decrypt) files from cloud storage</li>
<li>Mirror cloud data to other cloud services or locally</li>
<li>Migrate data to cloud, or between cloud storage vendors</li>
<li>Mount multiple, encrypted, cached or diverse cloud storage as a disk</li>
<li>Analyse and account for data held on cloud storage using <ahref="https://rclone.org/commands/rclone_lsf/">lsf</a>, <ahref="https://rclone.org/commands/rclone_lsjson/">ljson</a>, <ahref="https://rclone.org/commands/rclone_size/">size</a>, <ahref="https://rclone.org/commands/rclone_ncdu/">ncdu</a></li>
<li><ahref="https://rclone.org/union/">Union</a> file systems together to present multiple local and/or cloud file systems as one</li>
</ul>
<h2id="features">Features</h2>
<ul>
<li>Transfers
<ul>
<li>MD5, SHA1 hashes are checked at all times for file integrity</li>
<li>Can use multi-threaded downloads to local disk</li>
</ul></li>
<li><ahref="https://rclone.org/commands/rclone_copy/">Copy</a> new or changed files to cloud storage</li>
<li><ahref="https://rclone.org/commands/rclone_sync/">Sync</a> (one way) to make a directory identical</li>
<li><ahref="https://rclone.org/commands/rclone_move/">Move</a> files to cloud storage deleting the local after verification</li>
<li><ahref="https://rclone.org/commands/rclone_check/">Check</a> hashes and for missing/extra files</li>
<li><ahref="https://rclone.org/commands/rclone_mount/">Mount</a> your cloud storage as a network disk</li>
<li><ahref="https://rclone.org/commands/rclone_serve/">Serve</a> local or remote files over <ahref="https://rclone.org/commands/rclone_serve_http/">HTTP</a>/<ahref="https://rclone.org/commands/rclone_serve_webdav/">WebDav</a>/<ahref="https://rclone.org/commands/rclone_serve_ftp/">FTP</a>/<ahref="https://rclone.org/commands/rclone_serve_sftp/">SFTP</a>/<ahref="https://rclone.org/commands/rclone_serve_dlna/">dlna</a></li>
<li>Experimental <ahref="https://rclone.org/gui/">Web based GUI</a></li>
<p>Already installed rclone can be easily updated to the latest version using the <ahref="https://rclone.org/commands/rclone_selfupdate/">rclone selfupdate</a> command.</p>
<h2id="macos-installation-from-precompiled-binary-using-a-web-browser">macOS installation from precompiled binary, using a web browser</h2>
<p>When downloading a binary with a web browser, the browser will set the macOS gatekeeper quarantine attribute. Starting from Catalina, when attempting to run <code>rclone</code>, a pop-up will appear saying:</p>
<pre><code>“rclone” cannot be opened because the developer cannot be verified.
macOS cannot verify that this app is free from malware.</code></pre>
<h2id="install-with-docker">Install with docker</h2>
<p>The rclone maintains a <ahref="https://hub.docker.com/r/rclone/rclone">docker image for rclone</a>. These images are autobuilt by docker hub from the rclone source based on a minimal Alpine linux image.</p>
<p>The <code>:latest</code> tag will always point to the latest stable release. You can use the <code>:beta</code> tag to get the latest build from master. You can also use version tags, e.g. <code>:1.49.1</code>, <code>:1.49</code> or <code>:1</code>.</p>
<p>There are a few command line options to consider when starting an rclone Docker container from the rclone image.</p>
<ul>
<li><p>You need to mount the host rclone config dir at <code>/config/rclone</code> into the Docker container. Due to the fact that rclone updates tokens inside its config file, and that the update process involves a file rename, you need to mount the whole host rclone config dir, not just the single host rclone config file.</p></li>
<li><p>You need to mount a host data dir at <code>/data</code> into the Docker container.</p></li>
<li><p>By default, the rclone binary inside a Docker container runs with UID=0 (root). As a result, all files created in a run will have UID=0. If your config and data files reside on the host with a non-root UID:GID, you need to pass these on the container start command line.</p></li>
<li>If you want to access the RC interface (either via the API or the Web UI), it is required to set the <code>--rc-addr</code> to <code>:5572</code> in order to connect to it from outside the container. An explanation about why this is necessary is present <ahref="https://web.archive.org/web/20200808071950/https://pythonspeed.com/articles/docker-connection-refused/">here</a>.
<ul>
<li>NOTE: Users running this container with the docker network set to <code>host</code> should probably set it to listen to localhost only, with <code>127.0.0.1:5572</code> as the value for <code>--rc-addr</code></li>
</ul></li>
<li><p>It is possible to use <code>rclone mount</code> inside a userspace Docker container, and expose the resulting fuse mount to the host. The exact <code>docker run</code> options to do that might vary slightly between hosts. See, e.g. the discussion in this <ahref="https://github.com/moby/moby/issues/9448">thread</a>.</p>
<p>Make sure you have at least <ahref="https://golang.org/">Go</a> go1.13 installed. <ahref="https://golang.org/dl/">Download go</a> if necessary. The latest release is recommended. Then</p>
<p>This will leave you a checked out version of rclone you can modify and send pull requests with. If you use <code>make</code> instead of <code>go build</code> then the rclone build will have the correct version information in it.</p>
<p>You can also build the latest stable rclone with:</p>
<pre><code>go get github.com/rclone/rclone</code></pre>
<p>or the latest version (equivalent to the beta) with</p>
<pre><code>go get github.com/rclone/rclone@master</code></pre>
<p>These will build the binary in <code>$(go env GOPATH)/bin</code> (<code>~/go/bin/rclone</code> by default) after downloading the source to the go module cache. Note - do <strong>not</strong> use the <code>-u</code> flag here. This causes go to try to update the dependencies that rclone uses and sometimes these don't work with the current version of rclone.</p>
<p>First, you'll need to configure rclone. As the object storage systems have quite complicated authentication these are kept in a config file. (See the <ahref="#config-config-file"><code>--config</code></a> entry for how to find the config file and choose its location.)</p>
<p>Source and destination paths are specified by the name you gave the storage system in the config file then the sub path, e.g. "drive:myfolder" to look at "myfolder" in Google drive.</p>
<p>Enter an interactive configuration session where you can setup new remotes and manage existing ones. You may also set or remove a password to protect your configuration.</p>
<li><ahref="https://rclone.org/commands/rclone_config_dump/">rclone config dump</a> - Dump the config file as JSON.</li>
<li><ahref="https://rclone.org/commands/rclone_config_edit/">rclone config edit</a> - Enter an interactive configuration session.</li>
<li><ahref="https://rclone.org/commands/rclone_config_file/">rclone config file</a> - Show path of configuration file in use.</li>
<li><ahref="https://rclone.org/commands/rclone_config_password/">rclone config password</a> - Update password in an existing remote.</li>
<li><ahref="https://rclone.org/commands/rclone_config_providers/">rclone config providers</a> - List in JSON format all the providers and options.</li>
<li><ahref="https://rclone.org/commands/rclone_config_show/">rclone config show</a> - Print (decrypted) config file, or the config for a single remote.</li>
<li><ahref="https://rclone.org/commands/rclone_config_update/">rclone config update</a> - Update options in an existing remote.</li>
<p>Copy the source to the destination. Doesn't transfer unchanged files, testing by size and modification time or MD5SUM. Doesn't delete files from the destination.</p>
<p>Note that it is always the contents of the directory that is synced, not the directory so when source:path is a directory, it's the contents of source:path that are copied, not the directory name and contents.</p>
<p>If dest:path doesn't exist, it is created and the source:path contents go there.</p>
<p>If you are familiar with <code>rsync</code>, rclone always works as if you had written a trailing <code>/</code> - meaning "copy the contents of this directory". This applies to all commands and whether you are talking about the source or destination.</p>
<p>See the <ahref="https://rclone.org/docs/#no-traverse">--no-traverse</a> option for controlling whether rclone lists the destination directory or not. Supplying this option when copying a small number of files into a large destination can speed transfers up greatly.</p>
<p>For example, if you have many files in /path/to/src but only a few of them change every day, you can copy all the files which have changed recently very efficiently like this:</p>
<p>Sync the source to the destination, changing the destination only. Doesn't transfer unchanged files, testing by size and modification time or MD5SUM. Destination is updated to match source, including deleting files if necessary (except duplicate objects, see below).</p>
<p><strong>Important</strong>: Since this can cause data loss, test first with the <code>--dry-run</code> or the <code>--interactive</code>/<code>-i</code> flag.</p>
<p>Note that files in the destination won't be deleted if there were any errors at any point. Duplicate objects (files with the same name, on those providers that support it) are also not yet handled.</p>
<p>It is always the contents of the directory that is synced, not the directory so when source:path is a directory, it's the contents of source:path that are copied, not the directory name and contents. See extended explanation in the <code>copy</code> command above if unsure.</p>
<p>If dest:path doesn't exist, it is created and the source:path contents go there.</p>
<p><strong>Note</strong>: Use the <code>rclone dedupe</code> command to deal with "Duplicate object/directory found in source/destination - ignoring" errors. See <ahref="https://forum.rclone.org/t/sync-not-clearing-duplicates/14372">this forum post</a> for more info.</p>
<p>Moves the contents of the source directory to the destination directory. Rclone will error if the source and destination overlap and the remote does not support a server-side directory move operation.</p>
<p>If no filters are in use and if possible this will server-side move <code>source:path</code> into <code>dest:path</code>. After this <code>source:path</code> will no longer exist.</p>
<p>Otherwise for each file in <code>source:path</code> selected by the filters (if any) this will move it into <code>dest:path</code>. If possible a server-side move will be used, otherwise it will copy it (server-side if possible) into <code>dest:path</code> then delete the original (if no errors on copy) in <code>source:path</code>.</p>
<p>If you want to delete empty source directories after move, use the --delete-empty-src-dirs flag.</p>
<p>See the <ahref="https://rclone.org/docs/#no-traverse">--no-traverse</a> option for controlling whether rclone lists the destination directory or not. Supplying this option when moving a small number of files into a large destination can speed transfers up greatly.</p>
<p><strong>Important</strong>: Since this can cause data loss, test first with the <code>--dry-run</code> or the <code>--interactive</code>/<code>-i</code> flag.</p>
<p><code>rclone delete</code> only deletes files but leaves the directory structure alone. If you want to delete a directory and all of its contents use the <code>purge</code> command.</p>
<p>If you supply the <code>--rmdirs</code> flag, it will remove all empty directories along with it. You can also use the separate command <code>rmdir</code> or <code>rmdirs</code> to delete empty directories only.</p>
<p>For example, to delete all files bigger than 100MBytes, you may first want to check what would be deleted (use either):</p>
<p>That reads "delete everything with a minimum size of 100 MB", hence delete all files bigger than 100MBytes.</p>
<p><strong>Important</strong>: Since this can cause data loss, test first with the <code>--dry-run</code> or the <code>--interactive</code>/<code>-i</code> flag.</p>
<p>Remove the path and all of its contents. Note that this does not obey include/exclude filters - everything will be removed. Use the <code>delete</code> command if you want to selectively delete files. To delete empty directories only, use command <code>rmdir</code> or <code>rmdirs</code>.</p>
<p><strong>Important</strong>: Since this can cause data loss, test first with the <code>--dry-run</code> or the <code>--interactive</code>/<code>-i</code> flag.</p>
<p>This removes empty directory given by path. Will not remove the path if it has any objects in it, not even empty subdirectories. Use command <code>rmdirs</code> (or <code>delete</code> with option <code>--rmdirs</code>) to do that.</p>
<p>To delete a path and any objects in it, use <code>purge</code> command.</p>
<p>Checks the files in the source and destination match. It compares sizes and hashes (MD5 or SHA1) and logs a report of files which don't match. It doesn't alter the source or destination.</p>
<p>If you supply the <code>--size-only</code> flag, it will only compare the sizes not the hashes as well. Use this for a quick check.</p>
<p>If you supply the <code>--download</code> flag, it will download the data from both remotes and check them against each other on the fly. This can be useful for remotes that don't support hashes or if you really want to check all the data.</p>
<p>If you supply the <code>--one-way</code> flag, it will only check that files in the source match the files in the destination, not the other way around. This means that extra files in the destination that are not in the source will not be detected.</p>
<p>The <code>--differ</code>, <code>--missing-on-dst</code>, <code>--missing-on-src</code>, <code>--match</code> and <code>--error</code> flags write paths, one per line, to the file name (or stdout if it is <code>-</code>) supplied. What they write is described in the help below. For example <code>--differ</code> will write all paths which are present on both the source and destination but different.</p>
<p>The <code>--combined</code> flag will write a file (or stdout) which contains all file paths with a symbol and then a space and then the path to tell you what happened to it. These are reminiscent of diff files.</p>
<ul>
<li><code>= path</code> means path was found in source and destination and was identical</li>
<li><code>- path</code> means path was missing on the source, so only in the destination</li>
<li><code>+ path</code> means path was missing on the destination, so only in the source</li>
<li><code>* path</code> means path was present in source and destination but different.</li>
<li><code>! path</code> means there was an error reading or hashing the source or dest.</li>
<li><code>ls</code> to list size and path of objects only</li>
<li><code>lsl</code> to list modification time, size and path of objects only</li>
<li><code>lsd</code> to list directories only</li>
<li><code>lsf</code> to list objects and directories in easy to parse format</li>
<li><code>lsjson</code> to list objects and directories in JSON format</li>
</ul>
<p><code>ls</code>,<code>lsl</code>,<code>lsd</code> are designed to be human readable. <code>lsf</code> is designed to be human and machine readable. <code>lsjson</code> is designed to be machine readable.</p>
<p>Note that <code>ls</code> and <code>lsl</code> recurse by default - use <code>--max-depth 1</code> to stop the recursion.</p>
<p>The other list commands <code>lsd</code>,<code>lsf</code>,<code>lsjson</code> do not recurse by default - use <code>-R</code> to make them recurse.</p>
<p>Listing a non existent directory will produce an error except for remotes which can't have empty directories (e.g. s3, swift, or gcs - the bucket based remotes).</p>
<p>Lists the directories in the source path to standard output. Does not recurse by default. Use the -R flag to recurse.</p>
<p>This command lists the total size of the directory (if known, -1 if not), the modification time (if known, the current time if not), the number of objects in the directory (if known, -1 if not) and the name of the directory, Eg</p>
<li><code>ls</code> to list size and path of objects only</li>
<li><code>lsl</code> to list modification time, size and path of objects only</li>
<li><code>lsd</code> to list directories only</li>
<li><code>lsf</code> to list objects and directories in easy to parse format</li>
<li><code>lsjson</code> to list objects and directories in JSON format</li>
</ul>
<p><code>ls</code>,<code>lsl</code>,<code>lsd</code> are designed to be human readable. <code>lsf</code> is designed to be human and machine readable. <code>lsjson</code> is designed to be machine readable.</p>
<p>Note that <code>ls</code> and <code>lsl</code> recurse by default - use <code>--max-depth 1</code> to stop the recursion.</p>
<p>The other list commands <code>lsd</code>,<code>lsf</code>,<code>lsjson</code> do not recurse by default - use <code>-R</code> to make them recurse.</p>
<p>Listing a non existent directory will produce an error except for remotes which can't have empty directories (e.g. s3, swift, or gcs - the bucket based remotes).</p>
<li><code>ls</code> to list size and path of objects only</li>
<li><code>lsl</code> to list modification time, size and path of objects only</li>
<li><code>lsd</code> to list directories only</li>
<li><code>lsf</code> to list objects and directories in easy to parse format</li>
<li><code>lsjson</code> to list objects and directories in JSON format</li>
</ul>
<p><code>ls</code>,<code>lsl</code>,<code>lsd</code> are designed to be human readable. <code>lsf</code> is designed to be human and machine readable. <code>lsjson</code> is designed to be machine readable.</p>
<p>Note that <code>ls</code> and <code>lsl</code> recurse by default - use <code>--max-depth 1</code> to stop the recursion.</p>
<p>The other list commands <code>lsd</code>,<code>lsf</code>,<code>lsjson</code> do not recurse by default - use <code>-R</code> to make them recurse.</p>
<p>Listing a non existent directory will produce an error except for remotes which can't have empty directories (e.g. s3, swift, or gcs - the bucket based remotes).</p>
<p>By default, the hash is requested from the remote. If MD5 is not supported by the remote, no hash will be returned. With the download flag, the file will be downloaded from the remote and hashed locally enabling MD5 for any remote.</p>
<p>By default, the hash is requested from the remote. If SHA-1 is not supported by the remote, no hash will be returned. With the download flag, the file will be downloaded from the remote and hashed locally enabling SHA-1 for any remote.</p>
<p>Show the rclone version number, the go version, the build target OS and architecture, build tags and the type of executable (static or dynamic).</p>
<p>By default <code>dedupe</code> interactively finds files with duplicate names and offers to delete all but one or rename them to be different. This is known as deduping by name.</p>
<p>Deduping by name is only useful with a small group of backends (e.g. Google Drive, Opendrive) that can have duplicate file names. It can be run on wrapping backends (e.g. crypt) if they wrap a backend which supports duplicate file names.</p>
<p>However if --by-hash is passed in then dedupe will find files with duplicate hashes instead which will work on any backend which supports at least one hash. This can be used to find files with duplicate content. This is known as deduping by hash.</p>
<p>If deduping by name, first rclone will merge directories with the same name. It will do this iteratively until all the identically named directories have been merged.</p>
<p>Next, if deduping by name, for every group of duplicate file names / hashes, it will delete all but one identical files it finds without confirmation. This means that for most duplicated files the <code>dedupe</code> command will not be interactive.</p>
<p><code>dedupe</code> considers files to be identical if they have the same file path and the same hash. If the backend does not support hashes (e.g. crypt wrapping Google Drive) then they will never be found to be identical. If you use the <code>--size-only</code> flag then files will be considered identical if they have the same size (any hash will be ignored). This can be useful on crypt backends which do not support hashes.</p>
<p>Next rclone will resolve the remaining duplicates. Exactly which action is taken depends on the dedupe mode. By default rclone will interactively query the user for each one.</p>
<p><strong>Important</strong>: Since this can cause data loss, test first with the <code>--dry-run</code> or the <code>--interactive</code>/<code>-i</code> flag.</p>
<p><code>rclone about</code>prints quota information about a remote to standard output. The output is typically used, free, quota and trash contents.</p>
<p>E.g. Typical output from<code>rclone about remote:</code>is:</p>
<p>This runs a backend specific command. The commands themselves (except for "help" and "features") are defined by the backends and you should see the backend docs for definitions.</p>
<p>You can discover what commands a backend implements by using</p>
<pre><code>rclone backend help remote:
rclone backend help <backendname></code></pre>
<p>You can also discover information about the backend using (see <ahref="https://rclone.org/rc/#operations/fsinfo">operations/fsinfo</a> in the remote control docs for more info).</p>
<pre><code>rclone backend features remote:</code></pre>
<p>Use the <code>--head</code> flag to print characters only at the start, <code>--tail</code> for the end and <code>--offset</code> and <code>--count</code> to print a section in the middle. Note that if offset is negative it will count from the end, so <code>--offset -1 --count 1</code> is equivalent to <code>--tail 1</code>.</p>
<p>Create a new remote of <code>name</code> with <code>type</code> and options. The options should be passed in pairs of <code>key</code><code>value</code>.</p>
<p>Note that if the config process would normally ask a question the default is taken. Each time that happens rclone will print a message saying how to affect the value taken.</p>
<p>If any of the parameters passed is a password field, then rclone will automatically obscure them if they aren't already obscured before putting them in the config file.</p>
<p><strong>NB</strong> If the password parameter is 22 characters or longer and consists only of base64 characters then rclone can get confused about whether the password is already obscured or not and put unobscured passwords into the config file. If you want to be 100% certain that the passwords get obscured then use the "--obscure" flag, or if you are 100% certain you are already passing obscured passwords then use "--no-obscure". You can also set obscured passwords using the "rclone config password" command.</p>
<p>Enter an interactive configuration session where you can setup new remotes and manage existing ones. You may also set or remove a password to protect your configuration.</p>
<p>If any of the parameters passed is a password field, then rclone will automatically obscure them if they aren't already obscured before putting them in the config file.</p>
<p><strong>NB</strong> If the password parameter is 22 characters or longer and consists only of base64 characters then rclone can get confused about whether the password is already obscured or not and put unobscured passwords into the config file. If you want to be 100% certain that the passwords get obscured then use the "--obscure" flag, or if you are 100% certain you are already passing obscured passwords then use "--no-obscure". You can also set obscured passwords using the "rclone config password" command.</p>
<p>If source:path is a file or directory then it copies it to a file or directory named dest:path.</p>
<p>This can be used to upload single files to other than their current name. If the source is a directory then it acts exactly like the copy command.</p>
<p>So</p>
<pre><code>rclone copyto src dst</code></pre>
<p>where src and dst are rclone paths, either remote:path or /path/to/local or C:.</p>
<p>This will:</p>
<pre><code>if src is file
copy it to dst, overwriting an existing file if it exists
if src is directory
copy it to dst, overwriting existing files if they exist
<p>Setting <code>--auto-filename</code>will cause the file name to be retrieved from the from URL (after any redirections) and used in the destination path. With <code>--print-filename</code> in addition, the resuling file name will be printed.</p>
<p>Setting <code>--no-clobber</code> will prevent overwriting file on the destination if there is one with the same name.</p>
<p>Setting <code>--stdout</code> or making the output file name <code>-</code> will cause the output to be written to standard output.</p>
<p>rclone cryptcheck checks a remote against a crypted remote. This is the equivalent of running rclone check, but able to check the checksums of the crypted remote.</p>
<p>For it to work the underlying remote of the cryptedremote must support some kind of checksum.</p>
<p>It works by reading the nonce from each file on the cryptedremote: and using that to encrypt each file on the remote:. It then checks the checksum of the underlying file on the cryptedremote: against the checksum of the file it has just encrypted.</p>
<p>If you supply the <code>--one-way</code> flag, it will only check that files in the source match the files in the destination, not the other way around. This means that extra files in the destination that are not in the source will not be detected.</p>
<p>The <code>--differ</code>, <code>--missing-on-dst</code>, <code>--missing-on-src</code>, <code>--match</code> and <code>--error</code> flags write paths, one per line, to the file name (or stdout if it is <code>-</code>) supplied. What they write is described in the help below. For example <code>--differ</code> will write all paths which are present on both the source and destination but different.</p>
<p>The <code>--combined</code> flag will write a file (or stdout) which contains all file paths with a symbol and then a space and then the path to tell you what happened to it. These are reminiscent of diff files.</p>
<ul>
<li><code>= path</code> means path was found in source and destination and was identical</li>
<li><code>- path</code> means path was missing on the source, so only in the destination</li>
<li><code>+ path</code> means path was missing on the destination, so only in the source</li>
<li><code>* path</code> means path was present in source and destination but different.</li>
<li><code>! path</code> means there was an error reading or hashing the source or dest.</li>
<p>Another way to accomplish this is by using the <code>rclone backend encode</code> (or <code>decode</code>)command. See the documentation on the <code>crypt</code> overlay for more info.</p>
<p>Remove a single file from remote. Unlike <code>delete</code> it cannot be used to remove a directory and it doesn't obey include/exclude filters - if the specified file exists, it will always be removed.</p>
<p>This produces markdown docs for the rclone commands to the directory supplied. These are in a format suitable for hugo to render into the rclone.org website.</p>
<p>Produces a hash file for all the objects in the path using the hash named. The output is in the same format as the standard md5sum/sha1sum tool.</p>
<p>By default, the hash is requested from the remote. If the hash is not supported by the remote, no hash will be returned. With the download flag, the file will be downloaded from the remote and hashed locally enabling any hash for any remote.</p>
<p>Run without a hash to see the list of all supported hashes, e.g.</p>
rclone link --expire 1d remote:path/to/file</code></pre>
<p>If you supply the --expire flag, it will set the expiration time otherwise it will use the default (100 years). <strong>Note</strong> not all backends support the --expire flag - if the backend doesn't support it then the link returned won't expire.</p>
<p>Use the --unlink flag to remove existing public links to the file or folder. <strong>Note</strong> not all backends support "--unlink" flag - those that don't will just ignore it.</p>
<p>If successful, the last line of the output will contain the link. Exact capabilities depend on the remote, but the link will always by default be created with the least constraints – e.g. no expiry, no password protection, accessible without account.</p>
<p>List the contents of the source path (directories and objects) to standard output in a form which is easy to parse by scripts. By default this will just be the names of the objects and directories, one per line. The directories will have a / suffix.</p>
<p>If you specify "h" in the format you will get the MD5 hash by default, use the "--hash" flag to change which hash you want. Note that this can be returned as an empty string if it isn't available on the object (and for directories), "ERROR" if there was an error reading it from the object and "UNSUPPORTED" if that object does not support that hash type.</p>
<p>(Though "rclone md5sum ." is an easier way of typing this.)</p>
<p>By default the separator is ";" this can be changed with the --separator flag. Note that separators aren't escaped in the path so putting it last is a good strategy.</p>
<li><code>ls</code> to list size and path of objects only</li>
<li><code>lsl</code> to list modification time, size and path of objects only</li>
<li><code>lsd</code> to list directories only</li>
<li><code>lsf</code> to list objects and directories in easy to parse format</li>
<li><code>lsjson</code> to list objects and directories in JSON format</li>
</ul>
<p><code>ls</code>,<code>lsl</code>,<code>lsd</code> are designed to be human readable. <code>lsf</code> is designed to be human and machine readable. <code>lsjson</code> is designed to be machine readable.</p>
<p>Note that <code>ls</code> and <code>lsl</code> recurse by default - use <code>--max-depth 1</code> to stop the recursion.</p>
<p>The other list commands <code>lsd</code>,<code>lsf</code>,<code>lsjson</code> do not recurse by default - use <code>-R</code> to make them recurse.</p>
<p>Listing a non existent directory will produce an error except for remotes which can't have empty directories (e.g. s3, swift, or gcs - the bucket based remotes).</p>
<p>If --hash is not specified the Hashes property won't be emitted. The types of hash can be specified with the --hash-type parameter (which may be repeated). If --hash-type is set then it implies --hash.</p>
<p>If --no-modtime is specified then ModTime will be blank. This can speed things up on remotes where reading the ModTime takes an extra request (e.g. s3, swift).</p>
<p>If --no-mimetype is specified then MimeType will be blank. This can speed things up on remotes where reading the MimeType takes an extra request (e.g. s3, swift).</p>
<p>If --encrypted is not specified the Encrypted won't be emitted.</p>
<p>If --dirs-only is not specified files in addition to directories are returned</p>
<p>If --files-only is not specified directories in addition to the files will be returned.</p>
<p>The Path field will only show folders below the remote path being listed. If "remote:path" contains the file "subfolder/file.txt", the Path for "file.txt" will be "subfolder/file.txt", not "remote:path/subfolder/file.txt". When used without --recursive the Path will always be the same as Name.</p>
<p>If the directory is a bucket in a bucket based backend, then "IsBucket" will be set to true. This key won't be present unless it is "true".</p>
<p>The time is in RFC3339 format with up to nanosecond precision. The number of decimal digits in the seconds will depend on the precision that the remote can hold the times, so if times are accurate to the nearest millisecond (e.g. Google Drive) then 3 digits will always be shown ("2017-05-31T16:15:57.034+01:00") whereas if the times are accurate to the nearest second (Dropbox, Box, WebDav, etc.) no digits will be shown ("2017-05-31T16:15:57+01:00").</p>
<li><code>ls</code> to list size and path of objects only</li>
<li><code>lsl</code> to list modification time, size and path of objects only</li>
<li><code>lsd</code> to list directories only</li>
<li><code>lsf</code> to list objects and directories in easy to parse format</li>
<li><code>lsjson</code> to list objects and directories in JSON format</li>
</ul>
<p><code>ls</code>,<code>lsl</code>,<code>lsd</code> are designed to be human readable. <code>lsf</code> is designed to be human and machine readable. <code>lsjson</code> is designed to be machine readable.</p>
<p>Note that <code>ls</code> and <code>lsl</code> recurse by default - use <code>--max-depth 1</code> to stop the recursion.</p>
<p>The other list commands <code>lsd</code>,<code>lsf</code>,<code>lsjson</code> do not recurse by default - use <code>-R</code> to make them recurse.</p>
<p>Listing a non existent directory will produce an error except for remotes which can't have empty directories (e.g. s3, swift, or gcs - the bucket based remotes).</p>
<p>On Linux and OSX, you can either run mount in foreground mode or background (daemon) mode. Mount runs in foreground mode by default, use the <code>--daemon</code> flag to specify background mode. You can only run mount in foreground mode on Windows.</p>
<p>On Linux/macOS/FreeBSD start the mount like this, where <code>/path/to/local/mount</code> is an <strong>empty</strong><strong>existing</strong> directory:</p>
<p>On Windows you can start a mount in different ways. See <ahref="#mounting-modes-on-windows">below</a> for details. The following examples will mount to an automatically assigned drive, to specific drive letter <code>X:</code>, to path <code>C:\path\parent\mount</code> (where parent directory or drive must exist, and mount must <strong>not</strong> exist, and is not supported when <ahref="#mounting-modes-on-windows">mounting as a network drive</a>), and the last example will mount as network share <code>\\cloud\remote</code> and map it to an automatically assigned drive:</p>
rclone mount remote:path/to/files \\cloud\remote</code></pre>
<p>When the program ends while in foreground mode, either via Ctrl+C or receiving a SIGINT or SIGTERM signal, the mount should be automatically stopped.</p>
<p>When running in background mode the user will have to stop the mount manually:</p>
<p>The umount operation can fail, for example when the mountpoint is busy. When that happens, it is the user's responsibility to stop the mount manually.</p>
<p>The size of the mounted file system will be set according to information retrieved from the remote, the same as returned by the <ahref="https://rclone.org/commands/rclone_about/">rclone about</a> command. Remotes with unlimited storage may report the used size only, then an additional 1PB of free space is assumed. If the remote does not <ahref="https://rclone.org/overview/#optional-features">support</a> the about feature at all, then 1PB is set as both the total and the free size.</p>
<p><strong>Note</strong>: As of <code>rclone</code> 1.52.2, <code>rclone mount</code> now requires Go version 1.13 or newer on some platforms depending on the underlying FUSE library in use.</p>
<p><ahref="https://github.com/billziss-gh/winfsp">WinFsp</a> is an open source Windows File System Proxy which makes it easy to write user space file systems for Windows. It provides a FUSE emulation layer which rclone uses combination with <ahref="https://github.com/billziss-gh/cgofuse">cgofuse</a>. Both of these packages are by Bill Zissimopoulos who was very helpful during the implementation of rclone mount for Windows.</p>
<h3id="mounting-modes-on-windows">Mounting modes on windows</h3>
<p>Unlike other operating systems, Microsoft Windows provides a different filesystem type for network and fixed drives. It optimises access on the assumption fixed disk drives are fast and reliable, while network drives have relatively high latency and less reliability. Some settings can also be differentiated between the two types, for example that Windows Explorer should just display icons and not create preview thumbnails for image and video files on network drives.</p>
<p>In most cases, rclone will mount the remote as a normal, fixed disk drive by default. However, you can also choose to mount it as a remote network drive, often described as a network share. If you mount an rclone remote using the default, fixed drive mode and experience unexpected program errors, freezes or other issues, consider mounting as a network drive instead.</p>
<p>When mounting as a fixed disk drive you can either mount to an unused drive letter, or to a path representing a <strong>non-existent</strong> subdirectory of an <strong>existing</strong> parent directory or drive. Using the special value <code>*</code> will tell rclone to automatically assign the next available drive letter, starting with Z: and moving backward. Examples:</p>
<p>Option <code>--volname</code> can be used to set a custom volume name for the mounted file system. The default is to use the remote name and path.</p>
<p>To mount as network drive, you can add option <code>--network-mode</code> to your mount command. Mounting to a directory path is not supported in this mode, it is a limitation Windows imposes on junctions, so the remote must always be mounted to a drive letter.</p>
<pre><code>rclone mount remote:path/to/files X: --network-mode</code></pre>
<p>A volume name specified with <code>--volname</code> will be used to create the network share path. A complete UNC path, such as <code>\\cloud\remote</code>, optionally with path <code>\\cloud\remote\madeup\path</code>, will be used as is. Any other string will be used as the share part, after a default prefix <code>\\server\</code>. If no volume name is specified then <code>\\server\share</code> will be used. You must make sure the volume name is unique when you are mounting more than one drive, or else the mount command will fail. The share name will treated as the volume label for the mapped drive, shown in Windows Explorer etc, while the complete <code>\\server\share</code> will be reported as the remote UNC path by <code>net use</code> etc, just like a normal network drive mapping.</p>
<p>If you specify a full network share UNC path with <code>--volname</code>, this will implicitely set the <code>--network-mode</code> option, so the following two examples have same result:</p>
<pre><code>rclone mount remote:path/to/files X: --network-mode
rclone mount remote:path/to/files X: --volname \\server\share</code></pre>
<p>You may also specify the network share UNC path as the mountpoint itself. Then rclone will automatically assign a drive letter, same as with <code>*</code> and use that as mountpoint, and instead use the UNC path specified as the volume name, as if it were specified with the <code>--volname</code> option. This will also implicitely set the <code>--network-mode</code> option. This means the following two examples have same result:</p>
<pre><code>rclone mount remote:path/to/files \\cloud\remote
rclone mount remote:path/to/files * --volname \\cloud\remote</code></pre>
<p>There is yet another way to enable network mode, and to set the share path, and that is to pass the "native" libfuse/WinFsp option directly: <code>--fuse-flag --VolumePrefix=\server\share</code>. Note that the path must be with just a single backslash prefix in this case.</p>
<p><em>Note:</em> In previous versions of rclone this was the only supported method.</p>
<p><ahref="https://en.wikipedia.org/wiki/Drive_mapping">Read more about drive mapping</a></p>
<p>See also <ahref="#limitations">Limitations</a> section below.</p>
<p>The FUSE emulation layer on Windows must convert between the POSIX-based permission model used in FUSE, and the permission model used in Windows, based on access-control lists (ACL).</p>
<p>The mounted filesystem will normally get three entries in its access-control list (ACL), representing permissions for the POSIX permission scopes: Owner, group and others. By default, the owner and group will be taken from the current user, and the built-in group "Everyone" will be used to represent others. The user/group can be customized with FUSE options "UserName" and "GroupName", e.g. <code>-o UserName=user123 -o GroupName="Authenticated Users"</code>.</p>
<p>The permissions on each entry will be set according to <ahref="#options">options</a><code>--dir-perms</code> and <code>--file-perms</code>, which takes a value in traditional <ahref="https://en.wikipedia.org/wiki/File-system_permissions#Numeric_notation">numeric notation</a>, where the default corresponds to <code>--file-perms 0666 --dir-perms 0777</code>.</p>
<p>Note that the mapping of permissions is not always trivial, and the result you see in Windows Explorer may not be exactly like you expected. For example, when setting a value that includes write access, this will be mapped to individual permissions "write attributes", "write data" and "append data", but not "write extended attributes". Windows will then show this as basic permission "Special" instead of "Write", because "Write" includes the "write extended attributes" permission.</p>
<p>If you set POSIX permissions for only allowing access to the owner, using <code>--file-perms 0600 --dir-perms 0700</code>, the user group and the built-in "Everyone" group will still be given some special permissions, such as "read attributes" and "read permissions", in Windows. This is done for compatibility reasons, e.g. to allow users without additional permissions to be able to read basic metadata about files like in UNIX. One case that may arise is that other programs (incorrectly) interprets this as the file being accessible by everyone. For example an SSH client may warn about "unprotected private key file".</p>
<p>WinFsp 2021 (version 1.9, still in beta) introduces a new FUSE option "FileSecurity", that allows the complete specification of file security descriptors using <ahref="https://docs.microsoft.com/en-us/windows/win32/secauthz/security-descriptor-string-format">SDDL</a>. With this you can work around issues such as the mentioned "unprotected private key file" by specifying <code>-o FileSecurity="D:P(A;;FA;;;OW)"</code>, for file all access (FA) to the owner (OW).</p>
<p>Note that drives created as Administrator are not visible by other accounts (including the account that was elevated as Administrator). So if you start a Windows drive from an Administrative Command Prompt and then try to access the same drive from Explorer (which does not run as Administrator), you will not be able to see the new drive.</p>
<p>The easiest way around this is to start the drive from a normal command prompt. It is also possible to start a drive from the SYSTEM account (using <ahref="https://github.com/billziss-gh/winfsp/wiki/WinFsp-Service-Architecture">the WinFsp.Launcher infrastructure</a>) which creates drives accessible for everyone on the system or alternatively using <ahref="https://nssm.cc/usage">the nssm service manager</a>.</p>
<p>Without the use of <code>--vfs-cache-mode</code> this can only write files sequentially, it can only seek when reading. This means that many applications won't work with their files on an rclone mount without <code>--vfs-cache-mode writes</code> or <code>--vfs-cache-mode full</code>. See the <ahref="#vfs-file-caching">VFS File Caching</a> section for more info.</p>
<p>The bucket based remotes (e.g. Swift, S3, Google Compute Storage, B2, Hubic) do not support the concept of empty directories, so empty directories will have a tendency to disappear once they fall out of the directory cache.</p>
<p>File systems expect things to be 100% reliable, whereas cloud storage systems are a long way from 100% reliable. The rclone sync/copy commands cope with this with lots of retries. However rclone mount can't use retries in the same way without making local copies of the uploads. Look at the <ahref="#vfs-file-caching">VFS File Caching</a> for solutions to make mount more reliable.</p>
<p>You can use the flag <code>--attr-timeout</code> to set the time the kernel caches the attributes (size, modification time, etc.) for directory entries.</p>
<p>The default is <code>1s</code> which caches files just long enough to avoid too many callbacks to rclone from the kernel.</p>
<p>In theory 0s should be the correct value for filesystems which can change outside the control of the kernel. However this causes quite a few problems such as <ahref="https://github.com/rclone/rclone/issues/2157">rclone using too much memory</a>, <ahref="https://forum.rclone.org/t/rclone-1-39-vs-1-40-mount-issue/5112">rclone not serving files to samba</a> and <ahref="https://github.com/rclone/rclone/issues/2095#issuecomment-371141147">excessive time listing directories</a>.</p>
<p>The kernel can cache the info about a file for the time given by <code>--attr-timeout</code>. You may see corruption if the remote file changes length during this window. It will show up as either a truncated file or a file with garbage on the end. With <code>--attr-timeout 1s</code> this is very unlikely but not impossible. The higher you set <code>--attr-timeout</code> the more likely it is. The default setting of "1s" is the lowest setting which mitigates the problems above.</p>
<p>If you set it higher (<code>10s</code> or <code>1m</code> say) then the kernel will call back to rclone less often making it more efficient, however there is more chance of the corruption issue above.</p>
<p>When running rclone mount as a systemd service, it is possible to use Type=notify. In this case the service will enter the started state after the mountpoint has been successfully set up. Units having the rclone mount service specified as a requirement will see all files and folders immediately in this mode.</p>
<p><code>--vfs-read-chunk-size</code> will enable reading the source objects in parts. This can reduce the used download quota for some remotes by requesting only chunks from the remote that are actually read at the cost of an increased number of requests.</p>
<p>When <code>--vfs-read-chunk-size-limit</code> is also specified and greater than <code>--vfs-read-chunk-size</code>, the chunk size for each open file will get doubled for each chunk read, until the specified value is reached. A value of <code>-1</code> will disable the limit and the chunk size will grow indefinitely.</p>
<p>With <code>--vfs-read-chunk-size 100M</code> and <code>--vfs-read-chunk-size-limit 0</code> the following parts will be downloaded: 0-100M, 100M-200M, 200M-300M, 300M-400M and so on. When <code>--vfs-read-chunk-size-limit 500M</code> is specified, the result would be 0-100M, 100M-300M, 300M-700M, 700M-1200M, 1200M-1700M and so on.</p>
<p>This command uses the VFS layer. This adapts the cloud storage objects that rclone uses into something which looks much more like a disk filing system.</p>
<p>Cloud storage objects have lots of properties which aren't like disk files - you can't extend them or write to the middle of them, so the VFS layer has to deal with that. Because there is no one right way of doing this there are various options explained below.</p>
<p>The VFS layer also implements a directory cache - this caches info about files and directories (but not the data) in memory.</p>
<p>Using the <code>--dir-cache-time</code> flag, you can control how long a directory should be considered up to date and not refreshed from the backend. Changes made through the mount will appear immediately or invalidate the cache.</p>
<pre><code>--dir-cache-time duration Time to cache directory entries for. (default 5m0s)
--poll-interval duration Time to wait between polling for changes.</code></pre>
<p>However, changes made directly on the cloud storage by the web interface or a different copy of rclone will only be picked up once the directory cache expires if the backend configured does not support polling for changes. If the backend supports polling, changes will be picked up within the polling interval.</p>
<p>You can send a <code>SIGHUP</code> signal to rclone for it to flush all directory caches, regardless of how old they are. Assuming only one rclone instance is running, you can reset the cache like this:</p>
<p>Each open file will try to keep the specified amount of data in memory at all times. The buffered data is bound to one open file and won't be shared.</p>
<p>This flag is a upper limit for the used memory per open file. The buffer will only use memory for data that is downloaded but not not yet read. If the buffer is empty, only a small amount of memory will be used.</p>
<p>The maximum memory used by rclone for buffering can be up to <code>--buffer-size * open files</code>.</p>
<h2id="vfs-file-caching">VFS File Caching</h2>
<p>These flags control the VFS file caching options. File caching is necessary to make the VFS layer appear compatible with a normal file system. It can be disabled at the cost of some compatibility.</p>
<p>For example you'll need to enable VFS caching if you want to read and write simultaneously to a file. See below for more details.</p>
<p>Note that the VFS cache is separate from the cache backend and you may find that you need one or the other or both.</p>
<p>If run with <code>-vv</code> rclone will print the location of the file cache. The files are stored in the user cache file area which is OS dependent but can be controlled with <code>--cache-dir</code> or setting the appropriate environment variable.</p>
<p>The cache has 4 different modes selected by <code>--vfs-cache-mode</code>. The higher the cache mode the more compatible rclone becomes at the cost of using disk space.</p>
<p>Note that files are written back to the remote only when they are closed and if they haven't been accessed for --vfs-write-back second. If rclone is quit or dies with files that haven't been uploaded, these will be uploaded next time rclone is run with the same flags.</p>
<p>If using <code>--vfs-cache-max-size</code> note that the cache may exceed this size for two reasons. Firstly because it is only checked every <code>--vfs-cache-poll-interval</code>. Secondly because open files cannot be evicted from the cache.</p>
<p>You <strong>should not</strong> run two copies of rclone using the same VFS cache with the same or overlapping remotes if using <code>--vfs-cache-mode > off</code>. This can potentially cause data corruption if you do. You can work around this by giving each rclone its own cache hierarchy with <code>--cache-dir</code>. You don't need to worry about this if the remotes in use don't overlap.</p>
<p>This is very similar to "off" except that files opened for read AND write will be buffered to disk. This means that files opened for write will be a lot more compatible, but uses the minimal disk space.</p>
<p>So if an application only reads the starts of each file, then rclone will only buffer the start of the file. These files will appear to be their full size in the cache, but they will be sparse files with only the data that has been downloaded present in them.</p>
<p>This mode should support all normal file system operations and is otherwise identical to --vfs-cache-mode writes.</p>
<p>When reading a file rclone will read --buffer-size plus --vfs-read-ahead bytes ahead. The --buffer-size is buffered in memory whereas the --vfs-read-ahead is buffered on disk.</p>
<p>When using this mode it is recommended that --buffer-size is not set too big and --vfs-read-ahead is set large if required.</p>
<p><strong>IMPORTANT</strong> not all file systems support sparse files. In particular FAT/exFAT do not. Rclone will perform very badly if the cache directory is on a filesystem which doesn't support sparse files and it will log an ERROR message if one is detected.</p>
<p>These flags may be used to enable/disable features of the VFS for performance or other reasons.</p>
<p>In particular S3 and Swift benefit hugely from the --no-modtime flag (or use --use-server-modtime for a slightly different effect) as each read of the modification time takes a transaction.</p>
<pre><code>--no-checksum Don't compare checksums on up/download.
--no-modtime Don't read/write the modification time (can speed things up).
--no-seek Don't allow seeking in files.
--read-only Mount read-only.</code></pre>
<p>When rclone reads files from a remote it reads them in chunks. This means that rather than requesting the whole file rclone reads the chunk specified. This is advantageous because some cloud providers account for reads being all the data requested, not all the data delivered.</p>
<p>Rclone will keep doubling the chunk size requested starting at --vfs-read-chunk-size with a maximum of --vfs-read-chunk-size-limit unless it is set to "off" in which case there will be no limit.</p>
<pre><code>--vfs-read-chunk-size SizeSuffix Read the source objects in chunks. (default 128M)
--vfs-read-chunk-size-limit SizeSuffix Max chunk doubling size (default "off")</code></pre>
<p>Sometimes rclone is delivered reads or writes out of order. Rather than seeking rclone will wait a short time for the in sequence read or write to come in. These flags only come into effect when not using an on disk cache file.</p>
<pre><code>--vfs-read-wait duration Time to wait for in-sequence read before seeking. (default 20ms)
--vfs-write-wait duration Time to wait for in-sequence write before giving error. (default 1s)</code></pre>
<p>When using VFS write caching (--vfs-cache-mode with value writes or full), the global flag --transfers can be set to adjust the number of parallel uploads of modified files from cache (the related global flag --checkers have no effect on mount).</p>
<pre><code>--transfers int Number of file transfers to run in parallel. (default 4)</code></pre>
<p>File systems in modern Windows are case-insensitive but case-preserving: although existing files can be opened using any case, the exact case used to create the file is preserved and available for programs to query. It is not allowed for two files in the same directory to differ only by case.</p>
<p>The <code>--vfs-case-insensitive</code> mount flag controls how rclone handles these two cases. If its value is "false", rclone passes file names to the mounted file system as-is. If the flag is "true" (or appears without a value on command line), rclone may perform a "fixup" as explained below.</p>
<p>The user may specify a file name to open/delete/rename/etc with a case different than what is stored on mounted file system. If an argument refers to an existing file with exactly the same name, then the case of the existing file on the disk will be used. However, if a file name with exactly the same name is not found but a name differing only by case exists, rclone will transparently fixup the name. This fixup happens only when an existing file is requested. Case sensitivity of file names created anew by rclone is controlled by an underlying mounted file system.</p>
<p>Note that case sensitivity of the operating system running rclone (the target) may differ from case sensitivity of a file system mounted by rclone (the source). The flag controls whether "fixup" is performed to satisfy the target.</p>
<p>If the flag is not provided on the command line, then its default value depends on the operating system where rclone runs: "true" on Windows and macOS, "false" otherwise. If the flag is provided without a value, then it is "true".</p>
<h2id="alternate-report-of-used-bytes">Alternate report of used bytes</h2>
<p>Some backends, most notably S3, do not report the amount of bytes used. If you need this information to be available when running <code>df</code> on the filesystem, then pass the flag <code>--vfs-used-is-size</code> to rclone. With this flag set, instead of relying on the backend to report this information, rclone will scan the whole remote similar to <code>rclone size</code> and compute the total used space itself.</p>
<p><em>WARNING.</em> Contrary to <code>rclone size</code>, this flag ignores filters so that the result is accurate. However, this is very inefficient and may cost lots of API calls resulting in extra charges. Use it as a last resort and only with caching.</p>
-o, --option stringArray Option for libfuse/WinFsp. Repeat if required.
--poll-interval duration Time to wait between polling for changes. Must be smaller than dir-cache-time. Only on supported remotes. Set to 0 to disable. (default 1m0s)
--vfs-read-chunk-size SizeSuffix Read the source objects in chunks. (default 128M)
--vfs-read-chunk-size-limit SizeSuffix If greater than --vfs-read-chunk-size, double the chunk size after each chunk read, until the limit is reached. 'off' is unlimited. (default off)
--volname string Set the volume name. Supported on Windows and OSX only.
--write-back-cache Makes kernel buffer writes before sending them to rclone. Without this, writethrough caching is used. Not supported on Windows.</code></pre>
<p>This can be used to rename files or upload single files to other than their existing name. If the source is a directory then it acts exactly like the move command.</p>
<p>This doesn't transfer unchanged files, testing by size and modification time or MD5SUM. src will be deleted on successful transfer.</p>
<p><strong>Important</strong>: Since this can cause data loss, test first with the <code>--dry-run</code> or the <code>--interactive</code>/<code>-i</code> flag.</p>
<p>This displays a text based user interface allowing the navigation of a remote. It is most useful for answering the question - "What is using all my disk space?".</p>
<p>To make the user interface it first scans the entire remote given and builds an in memory representation. rclone ncdu can be used during this scanning phase and you will see it building up the directory structure as it goes along.</p>
<p>This an homage to the <ahref="https://dev.yorhel.nl/ncdu">ncdu tool</a> but for rclone remotes. It is missing lots of features at the moment but is useful as it stands.</p>
<p>In the rclone config file, human readable passwords are obscured. Obscuring them is done by encrypting them and writing them out in base64. This is <strong>not</strong> a secure way of encrypting these passwords as rclone can decrypt them - it is to prevent "eyedropping" - namely someone seeing a password in the rclone config file by accident.</p>
<p>Many equally important things (like access tokens) are not obscured in the config file. However it is very hard to shoulder surf a 64 character hex token.</p>
<p>This command can also accept a password through STDIN instead of an argument by passing a hyphen as an argument. This will use the first line of STDIN as the password not including the trailing newline.</p>
<p>If there is no data on STDIN to read, rclone obscure will default to obfuscating the hyphen itself.</p>
<p>If you want to encrypt the config file then please use config file encryption - see <ahref="https://rclone.org/commands/rclone_config/">rclone config</a> for more info.</p>
<p>This runs a command against a running rclone. Use the --url flag to specify an non default URL to connect on. This can be either a ":port" which is taken to mean "http://localhost:port" or a "host:port" which is taken to mean "http://host:port"</p>
<p>A username and password can be passed in with --user and --pass.</p>
<p>Note that --rc-addr, --rc-user, --rc-pass will be read also for --url, --user, --pass.</p>
<p>The --json parameter can be used to pass in a JSON blob as an input instead of key=value arguments. This is the only way of passing in more complicated values.</p>
<p>The -o/--opt option can be used to set a key "opt" with key, value options in the form "-o key=value" or "-o key". It can be repeated as many times as required. This is useful for rc commands which take the "opt" parameter which by convention is a dictionary of strings.</p>
<p>The -a/--arg option can be used to set strings in the "arg" value. It can be repeated as many times as required. This is useful for rc commands which take the "arg" parameter which by convention is a list of strings.</p>
<p>Use --loopback to connect to the rclone instance running "rclone rc". This is very useful for testing commands without having to run an rclone rc server, e.g.:</p>
<p>If the remote file already exists, it will be overwritten.</p>
<p>rcat will try to upload small files in a single request, which is usually more efficient than the streaming/chunked upload endpoints, which use multiple requests. Exact behaviour depends on the remote. What is considered a small file may be set through <code>--streaming-upload-cutoff</code>. Uploading only starts after the cutoff is reached or if the file ends before that. The data must fit into RAM. The cutoff needs to be small enough to adhere the limits of your remote, please see there. Generally speaking, setting this cutoff too high will decrease your performance.</p>
<p>Note that the upload can also not be retried because the data is not kept around until the upload succeeds. If you need to transfer a lot of data, you're better off caching locally and then <code>rclone move</code> it to the destination.</p>
<p>This is useful if you are controlling rclone via the rc API.</p>
<p>If you pass in a path to a directory, rclone will serve that directory for GET requests on the URL passed in. It will also open the URL in the browser when rclone is run.</p>
<p>See the <ahref="https://rclone.org/rc/">rc documentation</a> for more info on the rc flags.</p>
<pre><code>rclone rcd <path to files to serve>* [flags]</code></pre>
<p>This recursively removes any empty directories (including directories that only contain empty directories), that it finds under the path. The root path itself will also be removed if it is empty, unless you supply the <code>--leave-root</code> flag.</p>
<p>Use command <code>rmdir</code> to delete just the empty directory given by path, not recurse.</p>
<p>This is useful for tidying up remotes that rclone has left a lot of empty directories in. For example the <code>delete</code> command will delete files but leave the directory structure (unless used with option <code>--rmdirs</code>).</p>
<p>To delete a path and any objects in it, use <code>purge</code> command.</p>
<p>This command downloads the latest release of rclone and replaces the currently running binary. The download is verified with a hashsum and cryptographically signed signature.</p>
<p>If used without flags (or with implied <code>--stable</code> flag), this command will install the latest stable release. However, some issues may be fixed (or features added) only in the latest beta release. In such cases you should run the command with the <code>--beta</code> flag, i.e. <code>rclone selfupdate --beta</code>. You can check in advance what version would be installed by adding the <code>--check</code> flag, then repeat the command without it when you are satisfied.</p>
<p>Sometimes the rclone team may recommend you a concrete beta or stable rclone release to troubleshoot your issue or add a bleeding edge feature. The <code>--version VER</code> flag, if given, will update to the concrete version instead of the latest one. If you omit micro version from <code>VER</code> (for example <code>1.53</code>), the latest matching micro version will be used.</p>
<p>Upon successful update rclone will print a message that contains a previous version number. You will need it if you later decide to revert your update for some reason. Then you'll have to note the previous version and run the following command: <code>rclone selfupdate [--beta] OLDVER</code>. If the old version contains only dots and digits (for example <code>v1.54.0</code>) then it's a stable release so you won't need the <code>--beta</code> flag. Beta releases have an additional information similar to <code>v1.54.0-beta.5111.06f1c0c61</code>. (if you are a developer and use a locally built rclone, the version number will end with <code>-DEV</code>, you will have to rebuild it as it obvisously can't be distributed).</p>
<p>If you previously installed rclone via a package manager, the package may include local documentation or configure services. You may wish to update with the flag <code>--package deb</code> or <code>--package rpm</code> (whichever is correct for your OS) to update these too. This command with the default <code>--package zip</code> will update only the rclone executable so the local manual may become inaccurate after it.</p>
<p>The <code>rclone mount</code> command (https://rclone.org/commands/rclone_mount/) may or may not support extended FUSE options depending on the build and OS. <code>selfupdate</code> will refuse to update if the capability would be discarded.</p>
<p>Note: Windows forbids deletion of a currently running executable so this command will rename the old executable to 'rclone.old.exe' upon success.</p>
<p>Please note that this command was not available before rclone version 1.55. If it fails for you with the message <code>unknown command "selfupdate"</code> then you will need to update manually following the install instructions located at https://rclone.org/install/</p>
<pre><code>rclone selfupdate [flags]</code></pre>
<h2id="options-57">Options</h2>
<pre><code> --beta Install beta release.
--check Check for latest release, do not download.
-h, --help help for selfupdate
--output string Save the downloaded binary at a given path (default: replace running binary)
<p>rclone serve dlna is a DLNA media server for media stored in an rclone remote. Many devices, such as the Xbox and PlayStation, can automatically discover this server in the LAN and play audio/video from it. VLC is also supported. Service discovery uses UDP multicast packets (SSDP) and will thus only work on LANs.</p>
<p>Rclone will list all files present in the remote, without filtering based on media formats or file extensions. Additionally, there is no media transcoding support. This means that some players might show files that they are not able to play back correctly.</p>
<p>Use <code>--addr</code> to specify which IP address and port the server should listen on, e.g. <code>--addr 1.2.3.4:8000</code> or <code>--addr :8080</code> to listen to all IPs.</p>
<p>This command uses the VFS layer. This adapts the cloud storage objects that rclone uses into something which looks much more like a disk filing system.</p>
<p>Cloud storage objects have lots of properties which aren't like disk files - you can't extend them or write to the middle of them, so the VFS layer has to deal with that. Because there is no one right way of doing this there are various options explained below.</p>
<p>The VFS layer also implements a directory cache - this caches info about files and directories (but not the data) in memory.</p>
<p>Using the <code>--dir-cache-time</code> flag, you can control how long a directory should be considered up to date and not refreshed from the backend. Changes made through the mount will appear immediately or invalidate the cache.</p>
<pre><code>--dir-cache-time duration Time to cache directory entries for. (default 5m0s)
--poll-interval duration Time to wait between polling for changes.</code></pre>
<p>However, changes made directly on the cloud storage by the web interface or a different copy of rclone will only be picked up once the directory cache expires if the backend configured does not support polling for changes. If the backend supports polling, changes will be picked up within the polling interval.</p>
<p>You can send a <code>SIGHUP</code> signal to rclone for it to flush all directory caches, regardless of how old they are. Assuming only one rclone instance is running, you can reset the cache like this:</p>
<p>Each open file will try to keep the specified amount of data in memory at all times. The buffered data is bound to one open file and won't be shared.</p>
<p>This flag is a upper limit for the used memory per open file. The buffer will only use memory for data that is downloaded but not not yet read. If the buffer is empty, only a small amount of memory will be used.</p>
<p>The maximum memory used by rclone for buffering can be up to <code>--buffer-size * open files</code>.</p>
<h2id="vfs-file-caching-1">VFS File Caching</h2>
<p>These flags control the VFS file caching options. File caching is necessary to make the VFS layer appear compatible with a normal file system. It can be disabled at the cost of some compatibility.</p>
<p>For example you'll need to enable VFS caching if you want to read and write simultaneously to a file. See below for more details.</p>
<p>Note that the VFS cache is separate from the cache backend and you may find that you need one or the other or both.</p>
<p>If run with <code>-vv</code> rclone will print the location of the file cache. The files are stored in the user cache file area which is OS dependent but can be controlled with <code>--cache-dir</code> or setting the appropriate environment variable.</p>
<p>The cache has 4 different modes selected by <code>--vfs-cache-mode</code>. The higher the cache mode the more compatible rclone becomes at the cost of using disk space.</p>
<p>Note that files are written back to the remote only when they are closed and if they haven't been accessed for --vfs-write-back second. If rclone is quit or dies with files that haven't been uploaded, these will be uploaded next time rclone is run with the same flags.</p>
<p>If using <code>--vfs-cache-max-size</code> note that the cache may exceed this size for two reasons. Firstly because it is only checked every <code>--vfs-cache-poll-interval</code>. Secondly because open files cannot be evicted from the cache.</p>
<p>You <strong>should not</strong> run two copies of rclone using the same VFS cache with the same or overlapping remotes if using <code>--vfs-cache-mode > off</code>. This can potentially cause data corruption if you do. You can work around this by giving each rclone its own cache hierarchy with <code>--cache-dir</code>. You don't need to worry about this if the remotes in use don't overlap.</p>
<p>This is very similar to "off" except that files opened for read AND write will be buffered to disk. This means that files opened for write will be a lot more compatible, but uses the minimal disk space.</p>
<p>So if an application only reads the starts of each file, then rclone will only buffer the start of the file. These files will appear to be their full size in the cache, but they will be sparse files with only the data that has been downloaded present in them.</p>
<p>This mode should support all normal file system operations and is otherwise identical to --vfs-cache-mode writes.</p>
<p>When reading a file rclone will read --buffer-size plus --vfs-read-ahead bytes ahead. The --buffer-size is buffered in memory whereas the --vfs-read-ahead is buffered on disk.</p>
<p>When using this mode it is recommended that --buffer-size is not set too big and --vfs-read-ahead is set large if required.</p>
<p><strong>IMPORTANT</strong> not all file systems support sparse files. In particular FAT/exFAT do not. Rclone will perform very badly if the cache directory is on a filesystem which doesn't support sparse files and it will log an ERROR message if one is detected.</p>
<p>These flags may be used to enable/disable features of the VFS for performance or other reasons.</p>
<p>In particular S3 and Swift benefit hugely from the --no-modtime flag (or use --use-server-modtime for a slightly different effect) as each read of the modification time takes a transaction.</p>
<pre><code>--no-checksum Don't compare checksums on up/download.
--no-modtime Don't read/write the modification time (can speed things up).
--no-seek Don't allow seeking in files.
--read-only Mount read-only.</code></pre>
<p>When rclone reads files from a remote it reads them in chunks. This means that rather than requesting the whole file rclone reads the chunk specified. This is advantageous because some cloud providers account for reads being all the data requested, not all the data delivered.</p>
<p>Rclone will keep doubling the chunk size requested starting at --vfs-read-chunk-size with a maximum of --vfs-read-chunk-size-limit unless it is set to "off" in which case there will be no limit.</p>
<pre><code>--vfs-read-chunk-size SizeSuffix Read the source objects in chunks. (default 128M)
--vfs-read-chunk-size-limit SizeSuffix Max chunk doubling size (default "off")</code></pre>
<p>Sometimes rclone is delivered reads or writes out of order. Rather than seeking rclone will wait a short time for the in sequence read or write to come in. These flags only come into effect when not using an on disk cache file.</p>
<pre><code>--vfs-read-wait duration Time to wait for in-sequence read before seeking. (default 20ms)
--vfs-write-wait duration Time to wait for in-sequence write before giving error. (default 1s)</code></pre>
<p>When using VFS write caching (--vfs-cache-mode with value writes or full), the global flag --transfers can be set to adjust the number of parallel uploads of modified files from cache (the related global flag --checkers have no effect on mount).</p>
<pre><code>--transfers int Number of file transfers to run in parallel. (default 4)</code></pre>
<p>File systems in modern Windows are case-insensitive but case-preserving: although existing files can be opened using any case, the exact case used to create the file is preserved and available for programs to query. It is not allowed for two files in the same directory to differ only by case.</p>
<p>The <code>--vfs-case-insensitive</code> mount flag controls how rclone handles these two cases. If its value is "false", rclone passes file names to the mounted file system as-is. If the flag is "true" (or appears without a value on command line), rclone may perform a "fixup" as explained below.</p>
<p>The user may specify a file name to open/delete/rename/etc with a case different than what is stored on mounted file system. If an argument refers to an existing file with exactly the same name, then the case of the existing file on the disk will be used. However, if a file name with exactly the same name is not found but a name differing only by case exists, rclone will transparently fixup the name. This fixup happens only when an existing file is requested. Case sensitivity of file names created anew by rclone is controlled by an underlying mounted file system.</p>
<p>Note that case sensitivity of the operating system running rclone (the target) may differ from case sensitivity of a file system mounted by rclone (the source). The flag controls whether "fixup" is performed to satisfy the target.</p>
<p>If the flag is not provided on the command line, then its default value depends on the operating system where rclone runs: "true" on Windows and macOS, "false" otherwise. If the flag is provided without a value, then it is "true".</p>
<h2id="alternate-report-of-used-bytes-1">Alternate report of used bytes</h2>
<p>Some backends, most notably S3, do not report the amount of bytes used. If you need this information to be available when running <code>df</code> on the filesystem, then pass the flag <code>--vfs-used-is-size</code> to rclone. With this flag set, instead of relying on the backend to report this information, rclone will scan the whole remote similar to <code>rclone size</code> and compute the total used space itself.</p>
<p><em>WARNING.</em> Contrary to <code>rclone size</code>, this flag ignores filters so that the result is accurate. However, this is very inefficient and may cost lots of API calls resulting in extra charges. Use it as a last resort and only with caching.</p>
--no-checksum Don't compare checksums on up/download.
--no-modtime Don't read/write the modification time (can speed things up).
--no-seek Don't allow seeking in files.
--poll-interval duration Time to wait between polling for changes. Must be smaller than dir-cache-time. Only on supported remotes. Set to 0 to disable. (default 1m0s)
--vfs-read-chunk-size-limit SizeSuffix If greater than --vfs-read-chunk-size, double the chunk size after each chunk read, until the limit is reached. 'off' is unlimited. (default off)
--vfs-read-wait duration Time to wait for in-sequence read before seeking. (default 20ms)
<p>rclone serve ftp implements a basic ftp server to serve the remote over FTP protocol. This can be viewed with a ftp client or you can make a remote of type ftp to read and write it.</p>
<p>Use --addr to specify which IP address and port the server should listen on, e.g. --addr 1.2.3.4:8000 or --addr :8080 to listen to all IPs. By default it only listens on localhost. You can use port :0 to let the OS choose an available port.</p>
<p>This command uses the VFS layer. This adapts the cloud storage objects that rclone uses into something which looks much more like a disk filing system.</p>
<p>Cloud storage objects have lots of properties which aren't like disk files - you can't extend them or write to the middle of them, so the VFS layer has to deal with that. Because there is no one right way of doing this there are various options explained below.</p>
<p>The VFS layer also implements a directory cache - this caches info about files and directories (but not the data) in memory.</p>
<p>Using the <code>--dir-cache-time</code> flag, you can control how long a directory should be considered up to date and not refreshed from the backend. Changes made through the mount will appear immediately or invalidate the cache.</p>
<pre><code>--dir-cache-time duration Time to cache directory entries for. (default 5m0s)
--poll-interval duration Time to wait between polling for changes.</code></pre>
<p>However, changes made directly on the cloud storage by the web interface or a different copy of rclone will only be picked up once the directory cache expires if the backend configured does not support polling for changes. If the backend supports polling, changes will be picked up within the polling interval.</p>
<p>You can send a <code>SIGHUP</code> signal to rclone for it to flush all directory caches, regardless of how old they are. Assuming only one rclone instance is running, you can reset the cache like this:</p>
<p>Each open file will try to keep the specified amount of data in memory at all times. The buffered data is bound to one open file and won't be shared.</p>
<p>This flag is a upper limit for the used memory per open file. The buffer will only use memory for data that is downloaded but not not yet read. If the buffer is empty, only a small amount of memory will be used.</p>
<p>The maximum memory used by rclone for buffering can be up to <code>--buffer-size * open files</code>.</p>
<h2id="vfs-file-caching-2">VFS File Caching</h2>
<p>These flags control the VFS file caching options. File caching is necessary to make the VFS layer appear compatible with a normal file system. It can be disabled at the cost of some compatibility.</p>
<p>For example you'll need to enable VFS caching if you want to read and write simultaneously to a file. See below for more details.</p>
<p>Note that the VFS cache is separate from the cache backend and you may find that you need one or the other or both.</p>
<p>If run with <code>-vv</code> rclone will print the location of the file cache. The files are stored in the user cache file area which is OS dependent but can be controlled with <code>--cache-dir</code> or setting the appropriate environment variable.</p>
<p>The cache has 4 different modes selected by <code>--vfs-cache-mode</code>. The higher the cache mode the more compatible rclone becomes at the cost of using disk space.</p>
<p>Note that files are written back to the remote only when they are closed and if they haven't been accessed for --vfs-write-back second. If rclone is quit or dies with files that haven't been uploaded, these will be uploaded next time rclone is run with the same flags.</p>
<p>If using <code>--vfs-cache-max-size</code> note that the cache may exceed this size for two reasons. Firstly because it is only checked every <code>--vfs-cache-poll-interval</code>. Secondly because open files cannot be evicted from the cache.</p>
<p>You <strong>should not</strong> run two copies of rclone using the same VFS cache with the same or overlapping remotes if using <code>--vfs-cache-mode > off</code>. This can potentially cause data corruption if you do. You can work around this by giving each rclone its own cache hierarchy with <code>--cache-dir</code>. You don't need to worry about this if the remotes in use don't overlap.</p>
<p>This is very similar to "off" except that files opened for read AND write will be buffered to disk. This means that files opened for write will be a lot more compatible, but uses the minimal disk space.</p>
<p>So if an application only reads the starts of each file, then rclone will only buffer the start of the file. These files will appear to be their full size in the cache, but they will be sparse files with only the data that has been downloaded present in them.</p>
<p>This mode should support all normal file system operations and is otherwise identical to --vfs-cache-mode writes.</p>
<p>When reading a file rclone will read --buffer-size plus --vfs-read-ahead bytes ahead. The --buffer-size is buffered in memory whereas the --vfs-read-ahead is buffered on disk.</p>
<p>When using this mode it is recommended that --buffer-size is not set too big and --vfs-read-ahead is set large if required.</p>
<p><strong>IMPORTANT</strong> not all file systems support sparse files. In particular FAT/exFAT do not. Rclone will perform very badly if the cache directory is on a filesystem which doesn't support sparse files and it will log an ERROR message if one is detected.</p>
<p>These flags may be used to enable/disable features of the VFS for performance or other reasons.</p>
<p>In particular S3 and Swift benefit hugely from the --no-modtime flag (or use --use-server-modtime for a slightly different effect) as each read of the modification time takes a transaction.</p>
<pre><code>--no-checksum Don't compare checksums on up/download.
--no-modtime Don't read/write the modification time (can speed things up).
--no-seek Don't allow seeking in files.
--read-only Mount read-only.</code></pre>
<p>When rclone reads files from a remote it reads them in chunks. This means that rather than requesting the whole file rclone reads the chunk specified. This is advantageous because some cloud providers account for reads being all the data requested, not all the data delivered.</p>
<p>Rclone will keep doubling the chunk size requested starting at --vfs-read-chunk-size with a maximum of --vfs-read-chunk-size-limit unless it is set to "off" in which case there will be no limit.</p>
<pre><code>--vfs-read-chunk-size SizeSuffix Read the source objects in chunks. (default 128M)
--vfs-read-chunk-size-limit SizeSuffix Max chunk doubling size (default "off")</code></pre>
<p>Sometimes rclone is delivered reads or writes out of order. Rather than seeking rclone will wait a short time for the in sequence read or write to come in. These flags only come into effect when not using an on disk cache file.</p>
<pre><code>--vfs-read-wait duration Time to wait for in-sequence read before seeking. (default 20ms)
--vfs-write-wait duration Time to wait for in-sequence write before giving error. (default 1s)</code></pre>
<p>When using VFS write caching (--vfs-cache-mode with value writes or full), the global flag --transfers can be set to adjust the number of parallel uploads of modified files from cache (the related global flag --checkers have no effect on mount).</p>
<pre><code>--transfers int Number of file transfers to run in parallel. (default 4)</code></pre>
<p>File systems in modern Windows are case-insensitive but case-preserving: although existing files can be opened using any case, the exact case used to create the file is preserved and available for programs to query. It is not allowed for two files in the same directory to differ only by case.</p>
<p>The <code>--vfs-case-insensitive</code> mount flag controls how rclone handles these two cases. If its value is "false", rclone passes file names to the mounted file system as-is. If the flag is "true" (or appears without a value on command line), rclone may perform a "fixup" as explained below.</p>
<p>The user may specify a file name to open/delete/rename/etc with a case different than what is stored on mounted file system. If an argument refers to an existing file with exactly the same name, then the case of the existing file on the disk will be used. However, if a file name with exactly the same name is not found but a name differing only by case exists, rclone will transparently fixup the name. This fixup happens only when an existing file is requested. Case sensitivity of file names created anew by rclone is controlled by an underlying mounted file system.</p>
<p>Note that case sensitivity of the operating system running rclone (the target) may differ from case sensitivity of a file system mounted by rclone (the source). The flag controls whether "fixup" is performed to satisfy the target.</p>
<p>If the flag is not provided on the command line, then its default value depends on the operating system where rclone runs: "true" on Windows and macOS, "false" otherwise. If the flag is provided without a value, then it is "true".</p>
<h2id="alternate-report-of-used-bytes-2">Alternate report of used bytes</h2>
<p>Some backends, most notably S3, do not report the amount of bytes used. If you need this information to be available when running <code>df</code> on the filesystem, then pass the flag <code>--vfs-used-is-size</code> to rclone. With this flag set, instead of relying on the backend to report this information, rclone will scan the whole remote similar to <code>rclone size</code> and compute the total used space itself.</p>
<p><em>WARNING.</em> Contrary to <code>rclone size</code>, this flag ignores filters so that the result is accurate. However, this is very inefficient and may cost lots of API calls resulting in extra charges. Use it as a last resort and only with caching.</p>
<p>If you supply the parameter <code>--auth-proxy /path/to/program</code> then rclone will use that program to generate backends on the fly which then are used to authenticate incoming requests. This uses a simple JSON based protocol with input on STDIN and output on STDOUT.</p>
<p><strong>PLEASE NOTE:</strong><code>--auth-proxy</code> and <code>--authorized-keys</code> cannot be used together, if <code>--auth-proxy</code> is set the authorized keys option will be ignored.</p>
<p>There is an example program <ahref="https://github.com/rclone/rclone/blob/master/test_proxy.py">bin/test_proxy.py</a> in the rclone source code.</p>
<p>The program's job is to take a <code>user</code> and <code>pass</code> on the input and turn those into the config for a backend on STDOUT in JSON format. This config will have any default parameters for the backend added, but it won't use configuration from environment variables or command line options - it is the job of the proxy program to make a complete config.</p>
<p>This would mean that an SFTP backend would be created on the fly for the <code>user</code> and <code>pass</code>/<code>public_key</code> returned in the output to the host given. Note that since <code>_obscure</code> is set to <code>pass</code>, rclone will obscure the <code>pass</code> parameter before creating the backend (which is required for sftp backends).</p>
<p>The program can manipulate the supplied <code>user</code> in any way, for example to make proxy to many different sftp backends, you could make the <code>user</code> be <code>user@example.com</code> and then set the <code>host</code> to <code>example.com</code> in the output and the user to <code>user</code>. For security you'd probably want to restrict the <code>host</code> to a limited list.</p>
<p>Note that an internal cache is keyed on <code>user</code> so only use that for configuration, don't use <code>pass</code> or <code>public_key</code>. This also means that if a user's password or public-key is changed the cache will need to expire (which takes 5 mins) before it takes effect.</p>
--no-checksum Don't compare checksums on up/download.
--no-modtime Don't read/write the modification time (can speed things up).
--no-seek Don't allow seeking in files.
--pass string Password for authentication. (empty value allow every password)
--passive-port string Passive port range to use. (default "30000-32000")
--poll-interval duration Time to wait between polling for changes. Must be smaller than dir-cache-time. Only on supported remotes. Set to 0 to disable. (default 1m0s)
--vfs-read-chunk-size-limit SizeSuffix If greater than --vfs-read-chunk-size, double the chunk size after each chunk read, until the limit is reached. 'off' is unlimited. (default off)
--vfs-read-wait duration Time to wait for in-sequence read before seeking. (default 20ms)
<p>rclone serve http implements a basic web server to serve the remote over HTTP. This can be viewed in a web browser or you can make a remote of type http read from it.</p>
<p>Use --addr to specify which IP address and port the server should listen on, e.g. --addr 1.2.3.4:8000 or --addr :8080 to listen to all IPs. By default it only listens on localhost. You can use port :0 to let the OS choose an available port.</p>
<p>If you set --addr to listen on a public or LAN accessible IP address then using Authentication is advised - see the next section for info.</p>
<p>--server-read-timeout and --server-write-timeout can be used to control the timeouts on the server. Note that this is the total time for a transfer.</p>
<p>--max-header-bytes controls the maximum number of bytes the server will accept in the HTTP header.</p>
<p>--baseurl controls the URL prefix that rclone serves from. By default rclone will serve from the root. If you used --baseurl "/rclone" then rclone would serve from a URL starting with "/rclone/". This is useful if you wish to proxy rclone serve. Rclone automatically inserts leading and trailing "/" on --baseurl, so --baseurl "rclone", --baseurl "/rclone" and --baseurl "/rclone/" are all treated identically.</p>
<p>--template allows a user to specify a custom markup template for http and webdav serve functions. The server exports the following markup to be used within the template to server pages:</p>
<p>You can either use an htpasswd file which can take lots of users, or set a single username and password with the --user and --pass flags.</p>
<p>Use --htpasswd /path/to/htpasswd to provide an htpasswd file. This is in standard apache format and supports MD5, SHA1 and BCrypt for basic authentication. Bcrypt is recommended.</p>
<p>By default this will serve over http. If you want you can serve over https. You will need to supply the --cert and --key flags. If you wish to do client side certificate validation then you will need to supply --client-ca also.</p>
<p>--cert should be either a PEM encoded certificate or a concatenation of that with the CA certificate. --key should be the PEM encoded private key and --client-ca should be the PEM encoded client certificate authority certificate.</p>
<p>This command uses the VFS layer. This adapts the cloud storage objects that rclone uses into something which looks much more like a disk filing system.</p>
<p>Cloud storage objects have lots of properties which aren't like disk files - you can't extend them or write to the middle of them, so the VFS layer has to deal with that. Because there is no one right way of doing this there are various options explained below.</p>
<p>The VFS layer also implements a directory cache - this caches info about files and directories (but not the data) in memory.</p>
<p>Using the <code>--dir-cache-time</code> flag, you can control how long a directory should be considered up to date and not refreshed from the backend. Changes made through the mount will appear immediately or invalidate the cache.</p>
<pre><code>--dir-cache-time duration Time to cache directory entries for. (default 5m0s)
--poll-interval duration Time to wait between polling for changes.</code></pre>
<p>However, changes made directly on the cloud storage by the web interface or a different copy of rclone will only be picked up once the directory cache expires if the backend configured does not support polling for changes. If the backend supports polling, changes will be picked up within the polling interval.</p>
<p>You can send a <code>SIGHUP</code> signal to rclone for it to flush all directory caches, regardless of how old they are. Assuming only one rclone instance is running, you can reset the cache like this:</p>
<p>Each open file will try to keep the specified amount of data in memory at all times. The buffered data is bound to one open file and won't be shared.</p>
<p>This flag is a upper limit for the used memory per open file. The buffer will only use memory for data that is downloaded but not not yet read. If the buffer is empty, only a small amount of memory will be used.</p>
<p>The maximum memory used by rclone for buffering can be up to <code>--buffer-size * open files</code>.</p>
<h2id="vfs-file-caching-3">VFS File Caching</h2>
<p>These flags control the VFS file caching options. File caching is necessary to make the VFS layer appear compatible with a normal file system. It can be disabled at the cost of some compatibility.</p>
<p>For example you'll need to enable VFS caching if you want to read and write simultaneously to a file. See below for more details.</p>
<p>Note that the VFS cache is separate from the cache backend and you may find that you need one or the other or both.</p>
<p>If run with <code>-vv</code> rclone will print the location of the file cache. The files are stored in the user cache file area which is OS dependent but can be controlled with <code>--cache-dir</code> or setting the appropriate environment variable.</p>
<p>The cache has 4 different modes selected by <code>--vfs-cache-mode</code>. The higher the cache mode the more compatible rclone becomes at the cost of using disk space.</p>
<p>Note that files are written back to the remote only when they are closed and if they haven't been accessed for --vfs-write-back second. If rclone is quit or dies with files that haven't been uploaded, these will be uploaded next time rclone is run with the same flags.</p>
<p>If using <code>--vfs-cache-max-size</code> note that the cache may exceed this size for two reasons. Firstly because it is only checked every <code>--vfs-cache-poll-interval</code>. Secondly because open files cannot be evicted from the cache.</p>
<p>You <strong>should not</strong> run two copies of rclone using the same VFS cache with the same or overlapping remotes if using <code>--vfs-cache-mode > off</code>. This can potentially cause data corruption if you do. You can work around this by giving each rclone its own cache hierarchy with <code>--cache-dir</code>. You don't need to worry about this if the remotes in use don't overlap.</p>
<p>This is very similar to "off" except that files opened for read AND write will be buffered to disk. This means that files opened for write will be a lot more compatible, but uses the minimal disk space.</p>
<p>So if an application only reads the starts of each file, then rclone will only buffer the start of the file. These files will appear to be their full size in the cache, but they will be sparse files with only the data that has been downloaded present in them.</p>
<p>This mode should support all normal file system operations and is otherwise identical to --vfs-cache-mode writes.</p>
<p>When reading a file rclone will read --buffer-size plus --vfs-read-ahead bytes ahead. The --buffer-size is buffered in memory whereas the --vfs-read-ahead is buffered on disk.</p>
<p>When using this mode it is recommended that --buffer-size is not set too big and --vfs-read-ahead is set large if required.</p>
<p><strong>IMPORTANT</strong> not all file systems support sparse files. In particular FAT/exFAT do not. Rclone will perform very badly if the cache directory is on a filesystem which doesn't support sparse files and it will log an ERROR message if one is detected.</p>
<p>These flags may be used to enable/disable features of the VFS for performance or other reasons.</p>
<p>In particular S3 and Swift benefit hugely from the --no-modtime flag (or use --use-server-modtime for a slightly different effect) as each read of the modification time takes a transaction.</p>
<pre><code>--no-checksum Don't compare checksums on up/download.
--no-modtime Don't read/write the modification time (can speed things up).
--no-seek Don't allow seeking in files.
--read-only Mount read-only.</code></pre>
<p>When rclone reads files from a remote it reads them in chunks. This means that rather than requesting the whole file rclone reads the chunk specified. This is advantageous because some cloud providers account for reads being all the data requested, not all the data delivered.</p>
<p>Rclone will keep doubling the chunk size requested starting at --vfs-read-chunk-size with a maximum of --vfs-read-chunk-size-limit unless it is set to "off" in which case there will be no limit.</p>
<pre><code>--vfs-read-chunk-size SizeSuffix Read the source objects in chunks. (default 128M)
--vfs-read-chunk-size-limit SizeSuffix Max chunk doubling size (default "off")</code></pre>
<p>Sometimes rclone is delivered reads or writes out of order. Rather than seeking rclone will wait a short time for the in sequence read or write to come in. These flags only come into effect when not using an on disk cache file.</p>
<pre><code>--vfs-read-wait duration Time to wait for in-sequence read before seeking. (default 20ms)
--vfs-write-wait duration Time to wait for in-sequence write before giving error. (default 1s)</code></pre>
<p>When using VFS write caching (--vfs-cache-mode with value writes or full), the global flag --transfers can be set to adjust the number of parallel uploads of modified files from cache (the related global flag --checkers have no effect on mount).</p>
<pre><code>--transfers int Number of file transfers to run in parallel. (default 4)</code></pre>
<p>File systems in modern Windows are case-insensitive but case-preserving: although existing files can be opened using any case, the exact case used to create the file is preserved and available for programs to query. It is not allowed for two files in the same directory to differ only by case.</p>
<p>The <code>--vfs-case-insensitive</code> mount flag controls how rclone handles these two cases. If its value is "false", rclone passes file names to the mounted file system as-is. If the flag is "true" (or appears without a value on command line), rclone may perform a "fixup" as explained below.</p>
<p>The user may specify a file name to open/delete/rename/etc with a case different than what is stored on mounted file system. If an argument refers to an existing file with exactly the same name, then the case of the existing file on the disk will be used. However, if a file name with exactly the same name is not found but a name differing only by case exists, rclone will transparently fixup the name. This fixup happens only when an existing file is requested. Case sensitivity of file names created anew by rclone is controlled by an underlying mounted file system.</p>
<p>Note that case sensitivity of the operating system running rclone (the target) may differ from case sensitivity of a file system mounted by rclone (the source). The flag controls whether "fixup" is performed to satisfy the target.</p>
<p>If the flag is not provided on the command line, then its default value depends on the operating system where rclone runs: "true" on Windows and macOS, "false" otherwise. If the flag is provided without a value, then it is "true".</p>
<h2id="alternate-report-of-used-bytes-3">Alternate report of used bytes</h2>
<p>Some backends, most notably S3, do not report the amount of bytes used. If you need this information to be available when running <code>df</code> on the filesystem, then pass the flag <code>--vfs-used-is-size</code> to rclone. With this flag set, instead of relying on the backend to report this information, rclone will scan the whole remote similar to <code>rclone size</code> and compute the total used space itself.</p>
<p><em>WARNING.</em> Contrary to <code>rclone size</code>, this flag ignores filters so that the result is accurate. However, this is very inefficient and may cost lots of API calls resulting in extra charges. Use it as a last resort and only with caching.</p>
--htpasswd string htpasswd file - if not provided no authentication is done
--key string SSL PEM Private key
--max-header-bytes int Maximum size of request header (default 4096)
--no-checksum Don't compare checksums on up/download.
--no-modtime Don't read/write the modification time (can speed things up).
--no-seek Don't allow seeking in files.
--pass string Password for authentication.
--poll-interval duration Time to wait between polling for changes. Must be smaller than dir-cache-time. Only on supported remotes. Set to 0 to disable. (default 1m0s)
--read-only Mount read-only.
--realm string realm for authentication (default "rclone")
--server-read-timeout duration Timeout for server reading data (default 1h0m0s)
--server-write-timeout duration Timeout for server writing data (default 1h0m0s)
--vfs-read-chunk-size-limit SizeSuffix If greater than --vfs-read-chunk-size, double the chunk size after each chunk read, until the limit is reached. 'off' is unlimited. (default off)
--vfs-read-wait duration Time to wait for in-sequence read before seeking. (default 20ms)
<p>rclone serve restic implements restic's REST backend API over HTTP. This allows restic to use rclone as a data storage mechanism for cloud providers that restic does not support directly.</p>
<p>Once you have set up the remote, check it is working with, for example "rclone lsd remote:". You may have called the remote something other than "remote:" - just substitute whatever you called it in the following instructions.</p>
<p>Adding --cache-objects=false will cause rclone to stop caching objects returned from the List call. Caching is normally desirable as it speeds up downloading objects, saves transactions and uses very little memory.</p>
<p>Now you can <ahref="http://restic.readthedocs.io/en/latest/030_preparing_a_new_repo.html#rest-server">follow the restic instructions</a> on setting up restic.</p>
<p>Note that you will need restic 0.8.2 or later to interoperate with rclone.</p>
<p>Note that you can use the endpoint to host multiple repositories. Do this by adding a directory name or path after the URL. Note that these <strong>must</strong> end with /. Eg</p>
<p>Use --addr to specify which IP address and port the server should listen on, e.g. --addr 1.2.3.4:8000 or --addr :8080 to listen to all IPs. By default it only listens on localhost. You can use port :0 to let the OS choose an available port.</p>
<p>If you set --addr to listen on a public or LAN accessible IP address then using Authentication is advised - see the next section for info.</p>
<p>--server-read-timeout and --server-write-timeout can be used to control the timeouts on the server. Note that this is the total time for a transfer.</p>
<p>--max-header-bytes controls the maximum number of bytes the server will accept in the HTTP header.</p>
<p>--baseurl controls the URL prefix that rclone serves from. By default rclone will serve from the root. If you used --baseurl "/rclone" then rclone would serve from a URL starting with "/rclone/". This is useful if you wish to proxy rclone serve. Rclone automatically inserts leading and trailing "/" on --baseurl, so --baseurl "rclone", --baseurl "/rclone" and --baseurl "/rclone/" are all treated identically.</p>
<p>--template allows a user to specify a custom markup template for http and webdav serve functions. The server exports the following markup to be used within the template to server pages:</p>
<p>You can either use an htpasswd file which can take lots of users, or set a single username and password with the --user and --pass flags.</p>
<p>Use --htpasswd /path/to/htpasswd to provide an htpasswd file. This is in standard apache format and supports MD5, SHA1 and BCrypt for basic authentication. Bcrypt is recommended.</p>
<p>By default this will serve over http. If you want you can serve over https. You will need to supply the --cert and --key flags. If you wish to do client side certificate validation then you will need to supply --client-ca also.</p>
<p>--cert should be either a PEM encoded certificate or a concatenation of that with the CA certificate. --key should be the PEM encoded private key and --client-ca should be the PEM encoded client certificate authority certificate.</p>
<p>rclone serve sftp implements an SFTP server to serve the remote over SFTP. This can be used with an SFTP client or you can make a remote of type sftp to use with it.</p>
<p>--bwlimit will be respected for file transfers. Use --stats to control the stats printing.</p>
<p>You must provide some means of authentication, either with --user/--pass, an authorized keys file (specify location with --authorized-keys - the default is the same as ssh), an --auth-proxy, or set the --no-auth flag for no authentication when logging in.</p>
<p>Note that this also implements a small number of shell commands so that it can provide md5sum/sha1sum/df information for the rclone sftp backend. This means that is can support SHA1SUMs, MD5SUMs and the about command when paired with the rclone sftp backend.</p>
<p>This command uses the VFS layer. This adapts the cloud storage objects that rclone uses into something which looks much more like a disk filing system.</p>
<p>Cloud storage objects have lots of properties which aren't like disk files - you can't extend them or write to the middle of them, so the VFS layer has to deal with that. Because there is no one right way of doing this there are various options explained below.</p>
<p>The VFS layer also implements a directory cache - this caches info about files and directories (but not the data) in memory.</p>
<p>Using the <code>--dir-cache-time</code> flag, you can control how long a directory should be considered up to date and not refreshed from the backend. Changes made through the mount will appear immediately or invalidate the cache.</p>
<pre><code>--dir-cache-time duration Time to cache directory entries for. (default 5m0s)
--poll-interval duration Time to wait between polling for changes.</code></pre>
<p>However, changes made directly on the cloud storage by the web interface or a different copy of rclone will only be picked up once the directory cache expires if the backend configured does not support polling for changes. If the backend supports polling, changes will be picked up within the polling interval.</p>
<p>You can send a <code>SIGHUP</code> signal to rclone for it to flush all directory caches, regardless of how old they are. Assuming only one rclone instance is running, you can reset the cache like this:</p>
<p>Each open file will try to keep the specified amount of data in memory at all times. The buffered data is bound to one open file and won't be shared.</p>
<p>This flag is a upper limit for the used memory per open file. The buffer will only use memory for data that is downloaded but not not yet read. If the buffer is empty, only a small amount of memory will be used.</p>
<p>The maximum memory used by rclone for buffering can be up to <code>--buffer-size * open files</code>.</p>
<h2id="vfs-file-caching-4">VFS File Caching</h2>
<p>These flags control the VFS file caching options. File caching is necessary to make the VFS layer appear compatible with a normal file system. It can be disabled at the cost of some compatibility.</p>
<p>For example you'll need to enable VFS caching if you want to read and write simultaneously to a file. See below for more details.</p>
<p>Note that the VFS cache is separate from the cache backend and you may find that you need one or the other or both.</p>
<p>If run with <code>-vv</code> rclone will print the location of the file cache. The files are stored in the user cache file area which is OS dependent but can be controlled with <code>--cache-dir</code> or setting the appropriate environment variable.</p>
<p>The cache has 4 different modes selected by <code>--vfs-cache-mode</code>. The higher the cache mode the more compatible rclone becomes at the cost of using disk space.</p>
<p>Note that files are written back to the remote only when they are closed and if they haven't been accessed for --vfs-write-back second. If rclone is quit or dies with files that haven't been uploaded, these will be uploaded next time rclone is run with the same flags.</p>
<p>If using <code>--vfs-cache-max-size</code> note that the cache may exceed this size for two reasons. Firstly because it is only checked every <code>--vfs-cache-poll-interval</code>. Secondly because open files cannot be evicted from the cache.</p>
<p>You <strong>should not</strong> run two copies of rclone using the same VFS cache with the same or overlapping remotes if using <code>--vfs-cache-mode > off</code>. This can potentially cause data corruption if you do. You can work around this by giving each rclone its own cache hierarchy with <code>--cache-dir</code>. You don't need to worry about this if the remotes in use don't overlap.</p>
<p>This is very similar to "off" except that files opened for read AND write will be buffered to disk. This means that files opened for write will be a lot more compatible, but uses the minimal disk space.</p>
<p>So if an application only reads the starts of each file, then rclone will only buffer the start of the file. These files will appear to be their full size in the cache, but they will be sparse files with only the data that has been downloaded present in them.</p>
<p>This mode should support all normal file system operations and is otherwise identical to --vfs-cache-mode writes.</p>
<p>When reading a file rclone will read --buffer-size plus --vfs-read-ahead bytes ahead. The --buffer-size is buffered in memory whereas the --vfs-read-ahead is buffered on disk.</p>
<p>When using this mode it is recommended that --buffer-size is not set too big and --vfs-read-ahead is set large if required.</p>
<p><strong>IMPORTANT</strong> not all file systems support sparse files. In particular FAT/exFAT do not. Rclone will perform very badly if the cache directory is on a filesystem which doesn't support sparse files and it will log an ERROR message if one is detected.</p>
<p>These flags may be used to enable/disable features of the VFS for performance or other reasons.</p>
<p>In particular S3 and Swift benefit hugely from the --no-modtime flag (or use --use-server-modtime for a slightly different effect) as each read of the modification time takes a transaction.</p>
<pre><code>--no-checksum Don't compare checksums on up/download.
--no-modtime Don't read/write the modification time (can speed things up).
--no-seek Don't allow seeking in files.
--read-only Mount read-only.</code></pre>
<p>When rclone reads files from a remote it reads them in chunks. This means that rather than requesting the whole file rclone reads the chunk specified. This is advantageous because some cloud providers account for reads being all the data requested, not all the data delivered.</p>
<p>Rclone will keep doubling the chunk size requested starting at --vfs-read-chunk-size with a maximum of --vfs-read-chunk-size-limit unless it is set to "off" in which case there will be no limit.</p>
<pre><code>--vfs-read-chunk-size SizeSuffix Read the source objects in chunks. (default 128M)
--vfs-read-chunk-size-limit SizeSuffix Max chunk doubling size (default "off")</code></pre>
<p>Sometimes rclone is delivered reads or writes out of order. Rather than seeking rclone will wait a short time for the in sequence read or write to come in. These flags only come into effect when not using an on disk cache file.</p>
<pre><code>--vfs-read-wait duration Time to wait for in-sequence read before seeking. (default 20ms)
--vfs-write-wait duration Time to wait for in-sequence write before giving error. (default 1s)</code></pre>
<p>When using VFS write caching (--vfs-cache-mode with value writes or full), the global flag --transfers can be set to adjust the number of parallel uploads of modified files from cache (the related global flag --checkers have no effect on mount).</p>
<pre><code>--transfers int Number of file transfers to run in parallel. (default 4)</code></pre>
<p>File systems in modern Windows are case-insensitive but case-preserving: although existing files can be opened using any case, the exact case used to create the file is preserved and available for programs to query. It is not allowed for two files in the same directory to differ only by case.</p>
<p>The <code>--vfs-case-insensitive</code> mount flag controls how rclone handles these two cases. If its value is "false", rclone passes file names to the mounted file system as-is. If the flag is "true" (or appears without a value on command line), rclone may perform a "fixup" as explained below.</p>
<p>The user may specify a file name to open/delete/rename/etc with a case different than what is stored on mounted file system. If an argument refers to an existing file with exactly the same name, then the case of the existing file on the disk will be used. However, if a file name with exactly the same name is not found but a name differing only by case exists, rclone will transparently fixup the name. This fixup happens only when an existing file is requested. Case sensitivity of file names created anew by rclone is controlled by an underlying mounted file system.</p>
<p>Note that case sensitivity of the operating system running rclone (the target) may differ from case sensitivity of a file system mounted by rclone (the source). The flag controls whether "fixup" is performed to satisfy the target.</p>
<p>If the flag is not provided on the command line, then its default value depends on the operating system where rclone runs: "true" on Windows and macOS, "false" otherwise. If the flag is provided without a value, then it is "true".</p>
<h2id="alternate-report-of-used-bytes-4">Alternate report of used bytes</h2>
<p>Some backends, most notably S3, do not report the amount of bytes used. If you need this information to be available when running <code>df</code> on the filesystem, then pass the flag <code>--vfs-used-is-size</code> to rclone. With this flag set, instead of relying on the backend to report this information, rclone will scan the whole remote similar to <code>rclone size</code> and compute the total used space itself.</p>
<p><em>WARNING.</em> Contrary to <code>rclone size</code>, this flag ignores filters so that the result is accurate. However, this is very inefficient and may cost lots of API calls resulting in extra charges. Use it as a last resort and only with caching.</p>
<p>If you supply the parameter <code>--auth-proxy /path/to/program</code> then rclone will use that program to generate backends on the fly which then are used to authenticate incoming requests. This uses a simple JSON based protocol with input on STDIN and output on STDOUT.</p>
<p><strong>PLEASE NOTE:</strong><code>--auth-proxy</code> and <code>--authorized-keys</code> cannot be used together, if <code>--auth-proxy</code> is set the authorized keys option will be ignored.</p>
<p>There is an example program <ahref="https://github.com/rclone/rclone/blob/master/test_proxy.py">bin/test_proxy.py</a> in the rclone source code.</p>
<p>The program's job is to take a <code>user</code> and <code>pass</code> on the input and turn those into the config for a backend on STDOUT in JSON format. This config will have any default parameters for the backend added, but it won't use configuration from environment variables or command line options - it is the job of the proxy program to make a complete config.</p>
<p>This would mean that an SFTP backend would be created on the fly for the <code>user</code> and <code>pass</code>/<code>public_key</code> returned in the output to the host given. Note that since <code>_obscure</code> is set to <code>pass</code>, rclone will obscure the <code>pass</code> parameter before creating the backend (which is required for sftp backends).</p>
<p>The program can manipulate the supplied <code>user</code> in any way, for example to make proxy to many different sftp backends, you could make the <code>user</code> be <code>user@example.com</code> and then set the <code>host</code> to <code>example.com</code> in the output and the user to <code>user</code>. For security you'd probably want to restrict the <code>host</code> to a limited list.</p>
<p>Note that an internal cache is keyed on <code>user</code> so only use that for configuration, don't use <code>pass</code> or <code>public_key</code>. This also means that if a user's password or public-key is changed the cache will need to expire (which takes 5 mins) before it takes effect.</p>
--no-auth Allow connections with no authentication if set.
--no-checksum Don't compare checksums on up/download.
--no-modtime Don't read/write the modification time (can speed things up).
--no-seek Don't allow seeking in files.
--pass string Password for authentication.
--poll-interval duration Time to wait between polling for changes. Must be smaller than dir-cache-time. Only on supported remotes. Set to 0 to disable. (default 1m0s)
--vfs-read-chunk-size-limit SizeSuffix If greater than --vfs-read-chunk-size, double the chunk size after each chunk read, until the limit is reached. 'off' is unlimited. (default off)
--vfs-read-wait duration Time to wait for in-sequence read before seeking. (default 20ms)
<p>rclone serve webdav implements a basic webdav server to serve the remote over HTTP via the webdav protocol. This can be viewed with a webdav client, through a web browser, or you can make a remote of type webdav to read and write it.</p>
<p>If this flag is set to "auto" then rclone will choose the first supported hash on the backend or you can use a named hash such as "MD5" or "SHA-1".</p>
<p>Use --addr to specify which IP address and port the server should listen on, e.g. --addr 1.2.3.4:8000 or --addr :8080 to listen to all IPs. By default it only listens on localhost. You can use port :0 to let the OS choose an available port.</p>
<p>If you set --addr to listen on a public or LAN accessible IP address then using Authentication is advised - see the next section for info.</p>
<p>--server-read-timeout and --server-write-timeout can be used to control the timeouts on the server. Note that this is the total time for a transfer.</p>
<p>--max-header-bytes controls the maximum number of bytes the server will accept in the HTTP header.</p>
<p>--baseurl controls the URL prefix that rclone serves from. By default rclone will serve from the root. If you used --baseurl "/rclone" then rclone would serve from a URL starting with "/rclone/". This is useful if you wish to proxy rclone serve. Rclone automatically inserts leading and trailing "/" on --baseurl, so --baseurl "rclone", --baseurl "/rclone" and --baseurl "/rclone/" are all treated identically.</p>
<p>--template allows a user to specify a custom markup template for http and webdav serve functions. The server exports the following markup to be used within the template to server pages:</p>
<p>You can either use an htpasswd file which can take lots of users, or set a single username and password with the --user and --pass flags.</p>
<p>Use --htpasswd /path/to/htpasswd to provide an htpasswd file. This is in standard apache format and supports MD5, SHA1 and BCrypt for basic authentication. Bcrypt is recommended.</p>
<p>By default this will serve over http. If you want you can serve over https. You will need to supply the --cert and --key flags. If you wish to do client side certificate validation then you will need to supply --client-ca also.</p>
<p>--cert should be either a PEM encoded certificate or a concatenation of that with the CA certificate. --key should be the PEM encoded private key and --client-ca should be the PEM encoded client certificate authority certificate.</p>
<p>This command uses the VFS layer. This adapts the cloud storage objects that rclone uses into something which looks much more like a disk filing system.</p>
<p>Cloud storage objects have lots of properties which aren't like disk files - you can't extend them or write to the middle of them, so the VFS layer has to deal with that. Because there is no one right way of doing this there are various options explained below.</p>
<p>The VFS layer also implements a directory cache - this caches info about files and directories (but not the data) in memory.</p>
<p>Using the <code>--dir-cache-time</code> flag, you can control how long a directory should be considered up to date and not refreshed from the backend. Changes made through the mount will appear immediately or invalidate the cache.</p>
<pre><code>--dir-cache-time duration Time to cache directory entries for. (default 5m0s)
--poll-interval duration Time to wait between polling for changes.</code></pre>
<p>However, changes made directly on the cloud storage by the web interface or a different copy of rclone will only be picked up once the directory cache expires if the backend configured does not support polling for changes. If the backend supports polling, changes will be picked up within the polling interval.</p>
<p>You can send a <code>SIGHUP</code> signal to rclone for it to flush all directory caches, regardless of how old they are. Assuming only one rclone instance is running, you can reset the cache like this:</p>
<p>Each open file will try to keep the specified amount of data in memory at all times. The buffered data is bound to one open file and won't be shared.</p>
<p>This flag is a upper limit for the used memory per open file. The buffer will only use memory for data that is downloaded but not not yet read. If the buffer is empty, only a small amount of memory will be used.</p>
<p>The maximum memory used by rclone for buffering can be up to <code>--buffer-size * open files</code>.</p>
<h2id="vfs-file-caching-5">VFS File Caching</h2>
<p>These flags control the VFS file caching options. File caching is necessary to make the VFS layer appear compatible with a normal file system. It can be disabled at the cost of some compatibility.</p>
<p>For example you'll need to enable VFS caching if you want to read and write simultaneously to a file. See below for more details.</p>
<p>Note that the VFS cache is separate from the cache backend and you may find that you need one or the other or both.</p>
<p>If run with <code>-vv</code> rclone will print the location of the file cache. The files are stored in the user cache file area which is OS dependent but can be controlled with <code>--cache-dir</code> or setting the appropriate environment variable.</p>
<p>The cache has 4 different modes selected by <code>--vfs-cache-mode</code>. The higher the cache mode the more compatible rclone becomes at the cost of using disk space.</p>
<p>Note that files are written back to the remote only when they are closed and if they haven't been accessed for --vfs-write-back second. If rclone is quit or dies with files that haven't been uploaded, these will be uploaded next time rclone is run with the same flags.</p>
<p>If using <code>--vfs-cache-max-size</code> note that the cache may exceed this size for two reasons. Firstly because it is only checked every <code>--vfs-cache-poll-interval</code>. Secondly because open files cannot be evicted from the cache.</p>
<p>You <strong>should not</strong> run two copies of rclone using the same VFS cache with the same or overlapping remotes if using <code>--vfs-cache-mode > off</code>. This can potentially cause data corruption if you do. You can work around this by giving each rclone its own cache hierarchy with <code>--cache-dir</code>. You don't need to worry about this if the remotes in use don't overlap.</p>
<p>This is very similar to "off" except that files opened for read AND write will be buffered to disk. This means that files opened for write will be a lot more compatible, but uses the minimal disk space.</p>
<p>So if an application only reads the starts of each file, then rclone will only buffer the start of the file. These files will appear to be their full size in the cache, but they will be sparse files with only the data that has been downloaded present in them.</p>
<p>This mode should support all normal file system operations and is otherwise identical to --vfs-cache-mode writes.</p>
<p>When reading a file rclone will read --buffer-size plus --vfs-read-ahead bytes ahead. The --buffer-size is buffered in memory whereas the --vfs-read-ahead is buffered on disk.</p>
<p>When using this mode it is recommended that --buffer-size is not set too big and --vfs-read-ahead is set large if required.</p>
<p><strong>IMPORTANT</strong> not all file systems support sparse files. In particular FAT/exFAT do not. Rclone will perform very badly if the cache directory is on a filesystem which doesn't support sparse files and it will log an ERROR message if one is detected.</p>
<p>These flags may be used to enable/disable features of the VFS for performance or other reasons.</p>
<p>In particular S3 and Swift benefit hugely from the --no-modtime flag (or use --use-server-modtime for a slightly different effect) as each read of the modification time takes a transaction.</p>
<pre><code>--no-checksum Don't compare checksums on up/download.
--no-modtime Don't read/write the modification time (can speed things up).
--no-seek Don't allow seeking in files.
--read-only Mount read-only.</code></pre>
<p>When rclone reads files from a remote it reads them in chunks. This means that rather than requesting the whole file rclone reads the chunk specified. This is advantageous because some cloud providers account for reads being all the data requested, not all the data delivered.</p>
<p>Rclone will keep doubling the chunk size requested starting at --vfs-read-chunk-size with a maximum of --vfs-read-chunk-size-limit unless it is set to "off" in which case there will be no limit.</p>
<pre><code>--vfs-read-chunk-size SizeSuffix Read the source objects in chunks. (default 128M)
--vfs-read-chunk-size-limit SizeSuffix Max chunk doubling size (default "off")</code></pre>
<p>Sometimes rclone is delivered reads or writes out of order. Rather than seeking rclone will wait a short time for the in sequence read or write to come in. These flags only come into effect when not using an on disk cache file.</p>
<pre><code>--vfs-read-wait duration Time to wait for in-sequence read before seeking. (default 20ms)
--vfs-write-wait duration Time to wait for in-sequence write before giving error. (default 1s)</code></pre>
<p>When using VFS write caching (--vfs-cache-mode with value writes or full), the global flag --transfers can be set to adjust the number of parallel uploads of modified files from cache (the related global flag --checkers have no effect on mount).</p>
<pre><code>--transfers int Number of file transfers to run in parallel. (default 4)</code></pre>
<p>File systems in modern Windows are case-insensitive but case-preserving: although existing files can be opened using any case, the exact case used to create the file is preserved and available for programs to query. It is not allowed for two files in the same directory to differ only by case.</p>
<p>The <code>--vfs-case-insensitive</code> mount flag controls how rclone handles these two cases. If its value is "false", rclone passes file names to the mounted file system as-is. If the flag is "true" (or appears without a value on command line), rclone may perform a "fixup" as explained below.</p>
<p>The user may specify a file name to open/delete/rename/etc with a case different than what is stored on mounted file system. If an argument refers to an existing file with exactly the same name, then the case of the existing file on the disk will be used. However, if a file name with exactly the same name is not found but a name differing only by case exists, rclone will transparently fixup the name. This fixup happens only when an existing file is requested. Case sensitivity of file names created anew by rclone is controlled by an underlying mounted file system.</p>
<p>Note that case sensitivity of the operating system running rclone (the target) may differ from case sensitivity of a file system mounted by rclone (the source). The flag controls whether "fixup" is performed to satisfy the target.</p>
<p>If the flag is not provided on the command line, then its default value depends on the operating system where rclone runs: "true" on Windows and macOS, "false" otherwise. If the flag is provided without a value, then it is "true".</p>
<h2id="alternate-report-of-used-bytes-5">Alternate report of used bytes</h2>
<p>Some backends, most notably S3, do not report the amount of bytes used. If you need this information to be available when running <code>df</code> on the filesystem, then pass the flag <code>--vfs-used-is-size</code> to rclone. With this flag set, instead of relying on the backend to report this information, rclone will scan the whole remote similar to <code>rclone size</code> and compute the total used space itself.</p>
<p><em>WARNING.</em> Contrary to <code>rclone size</code>, this flag ignores filters so that the result is accurate. However, this is very inefficient and may cost lots of API calls resulting in extra charges. Use it as a last resort and only with caching.</p>
<p>If you supply the parameter <code>--auth-proxy /path/to/program</code> then rclone will use that program to generate backends on the fly which then are used to authenticate incoming requests. This uses a simple JSON based protocol with input on STDIN and output on STDOUT.</p>
<p><strong>PLEASE NOTE:</strong><code>--auth-proxy</code> and <code>--authorized-keys</code> cannot be used together, if <code>--auth-proxy</code> is set the authorized keys option will be ignored.</p>
<p>There is an example program <ahref="https://github.com/rclone/rclone/blob/master/test_proxy.py">bin/test_proxy.py</a> in the rclone source code.</p>
<p>The program's job is to take a <code>user</code> and <code>pass</code> on the input and turn those into the config for a backend on STDOUT in JSON format. This config will have any default parameters for the backend added, but it won't use configuration from environment variables or command line options - it is the job of the proxy program to make a complete config.</p>
<p>This would mean that an SFTP backend would be created on the fly for the <code>user</code> and <code>pass</code>/<code>public_key</code> returned in the output to the host given. Note that since <code>_obscure</code> is set to <code>pass</code>, rclone will obscure the <code>pass</code> parameter before creating the backend (which is required for sftp backends).</p>
<p>The program can manipulate the supplied <code>user</code> in any way, for example to make proxy to many different sftp backends, you could make the <code>user</code> be <code>user@example.com</code> and then set the <code>host</code> to <code>example.com</code> in the output and the user to <code>user</code>. For security you'd probably want to restrict the <code>host</code> to a limited list.</p>
<p>Note that an internal cache is keyed on <code>user</code> so only use that for configuration, don't use <code>pass</code> or <code>public_key</code>. This also means that if a user's password or public-key is changed the cache will need to expire (which takes 5 mins) before it takes effect.</p>
--htpasswd string htpasswd file - if not provided no authentication is done
--key string SSL PEM Private key
--max-header-bytes int Maximum size of request header (default 4096)
--no-checksum Don't compare checksums on up/download.
--no-modtime Don't read/write the modification time (can speed things up).
--no-seek Don't allow seeking in files.
--pass string Password for authentication.
--poll-interval duration Time to wait between polling for changes. Must be smaller than dir-cache-time. Only on supported remotes. Set to 0 to disable. (default 1m0s)
--read-only Mount read-only.
--realm string realm for authentication (default "rclone")
--server-read-timeout duration Timeout for server reading data (default 1h0m0s)
--server-write-timeout duration Timeout for server writing data (default 1h0m0s)
--vfs-read-chunk-size-limit SizeSuffix If greater than --vfs-read-chunk-size, double the chunk size after each chunk read, until the limit is reached. 'off' is unlimited. (default off)
--vfs-read-wait duration Time to wait for in-sequence read before seeking. (default 20ms)
<p>rclone settier changes storage tier or class at remote if supported. Few cloud storage services provides different storage classes on objects, for example AWS S3 and Glacier, Azure Blob storage - Hot, Cool and Archive, Google Cloud Storage, Regional Storage, Nearline, Coldline etc.</p>
<p>Note that, certain tier changes make objects not available to access immediately. For example tiering to archive in azure blob storage makes objects in frozen state, user can restore by setting tier to Hot/Cool, similarly S3 to Glacier makes object inaccessible.true</p>
<li><ahref="https://rclone.org/commands/rclone_test_histogram/">rclone test histogram</a> - Makes a histogram of file name characters.</li>
<li><ahref="https://rclone.org/commands/rclone_test_info/">rclone test info</a> - Discovers file name or other limitations for paths.</li>
<li><ahref="https://rclone.org/commands/rclone_test_makefiles/">rclone test makefiles</a> - Make a random file hierarchy in
<dir></li>
<li><ahref="https://rclone.org/commands/rclone_test_memory/">rclone test memory</a> - Load all the objects at remote:path into memory and report memory stats.</li>
</ul>
<h1id="rclone-test-histogram">rclone test histogram</h1>
<p>Makes a histogram of file name characters.</p>
<h2id="synopsis-60">Synopsis</h2>
<p>This command outputs JSON which shows the histogram of characters used in filenames in the remote:path specified.</p>
<p>The data doesn't contain any identifying information but is useful for the rclone developers when developing filename compression.</p>
<pre><code>rclone test histogram [remote:path] [flags]</code></pre>
<h2id="options-67">Options</h2>
<pre><code> -h, --help help for histogram</code></pre>
<p>See the <ahref="https://rclone.org/flags/">global flags page</a> for global options not listed here.</p>
<h2id="see-also-67">SEE ALSO</h2>
<ul>
<li><ahref="https://rclone.org/commands/rclone_test/">rclone test</a> - Run a test command</li>
</ul>
<h1id="rclone-test-info">rclone test info</h1>
<p>Discovers file name or other limitations for paths.</p>
<h2id="synopsis-61">Synopsis</h2>
<p>rclone info discovers what filenames and upload methods are possible to write to the paths passed in and how long they can be. It can take some time. It will write test files into the remote:path passed in. It outputs a bit of go code for each one.</p>
<p><strong>NB</strong> this can create undeletable files and other hazards - use with care</p>
<pre><code>rclone test info [remote:path]+ [flags]</code></pre>
<h2id="options-68">Options</h2>
<pre><code> --all Run all tests.
--check-control Check control characters.
--check-length Check max filename length.
--check-normalization Check UTF-8 Normalization.
--check-streaming Check uploads with indeterminate file size.
-h, --help help for info
--upload-wait duration Wait after writing a file.
--write-json string Write results to file.</code></pre>
<p>See the <ahref="https://rclone.org/flags/">global flags page</a> for global options not listed here.</p>
<h2id="see-also-68">SEE ALSO</h2>
<ul>
<li><ahref="https://rclone.org/commands/rclone_test/">rclone test</a> - Run a test command</li>
</ul>
<h1id="rclone-test-makefiles">rclone test makefiles</h1>
Make a random file hierarchy in
<dir>
<pre><code>rclone test makefiles <dir> [flags]</code></pre>
<h2id="options-69">Options</h2>
<pre><code> --files int Number of files to create (default 1000)
--files-per-directory int Average number of files per directory (default 10)
-h, --help help for makefiles
--max-file-size SizeSuffix Maximum size of files to create (default 100)
--max-name-length int Maximum size of file names (default 12)
--min-file-size SizeSuffix Minimum size of file to create
--min-name-length int Minimum size of file names (default 4)</code></pre>
<p>See the <ahref="https://rclone.org/flags/">global flags page</a> for global options not listed here.</p>
<h2id="see-also-69">SEE ALSO</h2>
<ul>
<li><ahref="https://rclone.org/commands/rclone_test/">rclone test</a> - Run a test command</li>
</ul>
<h1id="rclone-test-memory">rclone test memory</h1>
<p>Load all the objects at remote:path into memory and report memory stats.</p>
<pre><code>rclone test memory remote:path [flags]</code></pre>
<h2id="options-70">Options</h2>
<pre><code> -h, --help help for memory</code></pre>
<p>See the <ahref="https://rclone.org/flags/">global flags page</a> for global options not listed here.</p>
<h2id="see-also-70">SEE ALSO</h2>
<ul>
<li><ahref="https://rclone.org/commands/rclone_test/">rclone test</a> - Run a test command</li>
<p>The tree command has many options for controlling the listing which are compatible with the tree command. Note that not all of them have short options as they conflict with rclone's short options.</p>
<p>rclone normally syncs or copies directories. However, if the source remote points to a file, rclone will just copy that file. The destination remote must point to a directory - rclone will give the error <code>Failed to create file system for "remote:file": is a file not a directory</code> if it isn't.</p>
<p>It is recommended to use <code>copy</code> when copying individual files, not <code>sync</code>. They have pretty much the same effect but <code>copy</code> will use a lot less memory.</p>
<p>On Windows <code>\</code> may be used instead of <code>/</code> in local paths <strong>only</strong>, non local paths must use <code>/</code>. See <ahref="https://rclone.org/local/#windows-paths">local filesystem</a> documentation for more about Windows-specific paths.</p>
<p>This refers to a directory <code>path/to/dir</code> on <code>remote:</code> as defined in the config file (configured with <code>rclone config</code>).</p>
<p>On most backends this is refers to the same directory as <code>remote:path/to/dir</code> and that format should be preferred. On a very small number of remotes (FTP, SFTP, Dropbox for business) this will refer to a different directory. On these, paths without a leading <code>/</code> will refer to your "home" directory and paths with a leading <code>/</code> will refer to the root.</p>
<p>This is an advanced form for creating remotes on the fly. <code>backend</code> should be the name or prefix of a backend (the <code>type</code> in the config file) and all the configuration for the backend should be provided on the command line (or in environment variables).</p>
<p>To copy files and directories from <code>example.com</code> in the relative directory <code>path/to/dir</code> to <code>/tmp/dir</code> using sftp.</p>
<p>The above examples can also be written using a connection string syntax, so instead of providing the arguments as command line parameters <code>--http-url https://pub.rclone.org</code> they are provided as part of the remote specification as a kind of connection string.</p>
<p>These can apply to modify existing remotes as well as create new remotes with the on the fly syntax. This example is equivalent to adding the <code>--drive-shared-with-me</code> parameter to the remote <code>gdrive:</code>.</p>
<p>The major advantage to using the connection string style syntax is that it only applies the the remote, not to all the remotes of that type of the command line. A common confusion is this attempt to copy a file shared on google drive to the normal drive which <strong>does not work</strong> because the <code>--drive-shared-with-me</code> flag applies to both the source and the destination.</p>
<p>This will make <code>parameter</code> be <code>with"quote</code> and <code>parameter2</code> be <code>with'quote</code>.</p>
<p>If you leave off the <code>=parameter</code> then rclone will substitute <code>=true</code> which works very well with flags. For example to use s3 configured in the environment you could use:</p>
<p>Note that on the command line you might need to surround these connection strings with <code>"</code> or <code>'</code> to stop the shell interpreting any special characters within them.</p>
<p>If you are a shell master then you'll know which strings are OK and which aren't, but if you aren't sure then enclose them in <code>"</code> and use <code>'</code> as the inside quote. This syntax works on all OSes.</p>
<p>On Linux/macOS some characters are still interpreted inside <code>"</code> strings in the shell (notably <code>\</code> and <code>$</code> and <code>"</code>) so if your strings contain those you can swap the roles of <code>"</code> and <code>'</code> thus. (This syntax does not work on Windows.)</p>
<h4id="connection-strings-config-and-logging">Connection strings, config and logging</h4>
<p>If you supply extra configuration to a backend by command line flag, environment variable or connection string then rclone will add a suffix based on the hash of the config to the name of the remote, eg</p>
<h2id="quoting-and-the-shell">Quoting and the shell</h2>
<p>When you are typing commands to your computer you are using something called the command line shell. This interprets various characters in an OS specific way.</p>
<p>Here are some gotchas which may help users unfamiliar with the shell rules</p>
<p>If your names have spaces or shell metacharacters (e.g. <code>*</code>, <code>?</code>, <code>$</code>, <code>'</code>, <code>"</code>, etc.) then you must quote them. Use single quotes <code>'</code> by default.</p>
<p>If you are using the root directory on its own then don't quote it (see <ahref="https://github.com/rclone/rclone/issues/464">#464</a> for why), e.g.</p>
<h2id="copying-files-or-directories-with-in-the-names">Copying files or directories with <code>:</code> in the names</h2>
<p>rclone uses <code>:</code> to mark a remote name. This is, however, a valid filename component in non-Windows OSes. The remote name parser will only search for a <code>:</code> up to the first <code>/</code> so if you need to act on a file or directory like this then use the full path starting with a <code>/</code>, or use <code>./</code> as a current directory prefix.</p>
<p>So to sync a directory called <code>sync:me</code> to a remote called <code>remote:</code> use</p>
<p>This means if you want to copy one folder to another then rclone won't download all the files and re-upload them; it will instruct the server to copy them in place.</p>
<p>Remotes which don't support server-side copy <strong>will</strong> download and re-upload in this case.</p>
<p>Server side copies are used with <code>sync</code> and <code>copy</code> and will be identified in the log when using the <code>-v</code> flag. The <code>move</code> command may also use them if remote doesn't support server-side move directly. This is done by issuing a server-side copy then a delete which is much quicker than a download and re-upload.</p>
<p>Options that take parameters can have the values passed in two ways, <code>--option=value</code> or <code>--option value</code>. However boolean (true/false) options behave slightly differently to the other options in that <code>--boolean</code> sets the option to <code>true</code> and the absence of the flag sets it to <code>false</code>. It is also possible to specify <code>--boolean=false</code> or <code>--boolean=true</code>. Note that <code>--boolean false</code> is not valid - this is parsed as <code>--boolean</code> and the <code>false</code> is parsed as an extra command line argument for rclone.</p>
<p>Options which use TIME use the go time parser. A duration string is a possibly signed sequence of decimal numbers, each with optional fraction and a unit suffix, such as "300ms", "-1.5h" or "2h45m". Valid time units are "ns", "us" (or "µs"), "ms", "s", "m", "h".</p>
<p>Options which use SIZE use kByte by default. However, a suffix of <code>b</code> for bytes, <code>k</code> for kBytes, <code>M</code> for MBytes, <code>G</code> for GBytes, <code>T</code> for TBytes and <code>P</code> for PBytes may be used. These are the binary units, e.g. 1, 2**10, 2**20, 2**30 respectively.</p>
<p>When using <code>sync</code>, <code>copy</code> or <code>move</code> any files which would have been overwritten or deleted are moved in their original hierarchy into this directory.</p>
<p>If <code>--suffix</code> is set, then the moved files will have the suffix added to them. If there is a file with the same path (after the suffix has been added) in DIR, then it will be overwritten.</p>
<p>The remote in use must support server-side move or copy and you must use the same remote as the destination of the sync. The backup directory must not overlap the destination directory.</p>
<p>will sync <code>/path/to/local</code> to <code>remote:current</code>, but for any files which would have been updated or deleted will be stored in <code>remote:old</code>.</p>
<p>If running rclone from a script you might want to use today's date as the directory name passed to <code>--backup-dir</code> to store the old files, or you might want to pass <code>--suffix</code> with today's date.</p>
<p>Local address to bind to for outgoing connections. This can be an IPv4 address (1.2.3.4), an IPv6 address (1234::789A) or host name. If the host name doesn't resolve or resolves to more than one IP address it will give an error.</p>
<p>would mean limit the upload and download bandwidth to 10 MByte/s. <strong>NB</strong> this is <strong>bytes</strong> per second not <strong>bits</strong> per second. To use a single limit, specify the desired bandwidth in kBytes/s, or use a suffix b|k|M|G. The default is <code>0</code> which means to not limit bandwidth.</p>
<p>would mean limit the upload bandwidth to 10 MByte/s and the download bandwidth to 100 kByte/s. Either limit can be "off" meaning no limit, so to just limit the upload bandwidth you would use</p>
<p>this would limit the upload bandwidth to 10MByte/s but the download bandwidth would be unlimited.</p>
<p>When specified as above the bandwidth limits last for the duration of run of the rclone binary.</p>
<p>It is also possible to specify a "timetable" of limits, which will cause certain limits to be applied at certain times. To specify a timetable, format your entries as <code>WEEKDAY-HH:MM,BANDWIDTH WEEKDAY-HH:MM,BANDWIDTH...</code> where: <code>WEEKDAY</code> is optional element.</p>
<ul>
<li><code>BANDWIDTH</code> can be a single number, e.g.<code>100k</code> or a pair of numbers for upload:download, e.g.<code>10M:1M</code>.</li>
<li><code>WEEKDAY</code> can be written as the whole word or only using the first 3 characters. It is optional.</li>
<li><code>HH:MM</code> is an hour from 00:00 to 23:59.</li>
<p>In this example, the transfer bandwidth will be set to 512kBytes/sec at 8am every day. At noon, it will rise to 10MByte/s, and drop back to 512kBytes/sec at 1pm. At 6pm, the bandwidth limit will be set to 30MByte/s, and at 11pm it will be completely disabled (full speed). Anything between 11pm and 8am will remain unlimited.</p>
<p>It means that, the transfer bandwidth will be set to 512kBytes/sec on Monday. It will rise to 10MByte/s before the end of Friday. At 10:00 on Saturday it will be set to 1MByte/s. From 20:00 on Sunday it will be unlimited.</p>
<p>Bandwidth limit apply to the data transfer for all backends. For most backends the directory listing bandwidth is also included (exceptions being the non HTTP backends, <code>ftp</code>, <code>sftp</code> and <code>tardigrade</code>).</p>
<p>Note that the units are <strong>Bytes/s</strong>, not <strong>Bits/s</strong>. Typically connections are measured in Bits/s - to convert divide by 8. For example, let's say you have a 10 Mbit/s connection and you wish rclone to use half of it - 5 Mbit/s. This is 5/8 = 0.625MByte/s so you would use a <code>--bwlimit 0.625M</code> parameter for rclone.</p>
<p>On Unix systems (Linux, macOS, …) the bandwidth limiter can be toggled by sending a <code>SIGUSR2</code> signal to rclone. This allows to remove the limitations of a long running rclone transfer and to restore it back to the value specified with <code>--bwlimit</code> quickly when needed. Assuming there is only one rclone instance running, you can toggle the limiter like this:</p>
<p>When using <code>mount</code> or <code>cmount</code> each open file descriptor will use this much memory for buffering. See the <ahref="https://rclone.org/commands/rclone_mount/#file-buffering">mount</a> documentation for more details.</p>
<p>Set to <code>0</code> to disable the buffering for the minimum memory usage.</p>
<p>If this flag is set then in a <code>sync</code>, <code>copy</code> or <code>move</code>, rclone will do all the checks to see whether files need to be transferred before doing any of the transfers. Normally rclone would start running transfers as soon as possible.</p>
<p>This flag can be useful on IO limited systems where transfers interfere with checking.</p>
<p>Using this flag can use more memory as it effectively sets <code>--max-backlog</code> to infinite. This means that all the info on the objects to transfer is held in memory before the transfers start.</p>
<p>The number of checkers to run in parallel. Checkers do the equality checking of files during a sync. For some storage systems (e.g. S3, Swift, Dropbox) this can take a significant amount of time so they are run in parallel.</p>
<p>Normally rclone will look at modification time and size of files to see if they are equal. If you set this flag then rclone will check the file hash and size to determine if files are equal.</p>
<p>This is very useful when transferring between remotes which store the same hash type on the object, e.g. Drive and Swift. For details of which remotes support which hash type see the table in the <ahref="https://rclone.org/overview/">overview section</a>.</p>
<p>When using <code>sync</code>, <code>copy</code> or <code>move</code> DIR is checked in addition to the destination for files. If a file identical to the source is found that file is NOT copied from source. This is useful to copy just files that have changed since the last backup.</p>
<p>You must use the same remote as the destination of the sync. The compare directory must not overlap the destination directory.</p>
<p>See <code>--copy-dest</code> and <code>--backup-dir</code>.</p>
<p>Normally the config file is in your home directory as a file called <code>.config/rclone/rclone.conf</code> (or <code>.rclone.conf</code> if created with an older version). If <code>$XDG_CONFIG_HOME</code> is set it will be at <code>$XDG_CONFIG_HOME/rclone/rclone.conf</code>.</p>
<p>If there is a file <code>rclone.conf</code> in the same directory as the rclone executable it will be preferred. This file must be created manually for Rclone to use it, it will never be created automatically.</p>
<p>If the location is set to empty string <code>""</code> or the special value <code>/notfound</code>, or the os null device represented by value <code>NUL</code> on Windows and <code>/dev/null</code> on Unix systems, then rclone will keep the config file in memory only.</p>
<p>The file format is basic <ahref="https://en.wikipedia.org/wiki/INI_file#Format">INI</a>: Sections of text, led by a <code>[section]</code> header and followed by <code>key=value</code> entries on separate lines. In rclone each remote is represented by its own section, where the section name defines the name of the remote. Options are specified as the <code>key=value</code> entries, where the key is the option name without the <code>--backend-</code> prefix, in lowercase and with <code>_</code> instead of <code>-</code>. E.g. option <code>--mega-hard-delete</code> corresponds to key <code>hard_delete</code>. Only backend options can be specified. A special, and required, key <code>type</code> identifies the <ahref="https://rclone.org/overview/">storage system</a>, where the value is the internal lowercase name as returned by command <code>rclone help backends</code>. Comments are indicated by <code>;</code> or <code>#</code> at the beginning of a line.</p>
<p>Note that passwords are in <ahref="https://rclone.org/commands/rclone_obscure/">obscured</a> form. Also, many storage systems uses token-based authentication instead of passwords, and this requires additional steps. It is easier, and safer, to use the interactive command <code>rclone config</code> instead of manually editing the configuration file.</p>
<p>The configuration file will typically contain login information, and should therefore have restricted permissions so that only the current user can read it. Rclone tries to ensure this when it writes the file. You may also choose to <ahref="#configuration-encryption">encrypt</a> the file.</p>
<p>When token-based authentication are used, the configuration file must be writable, because rclone needs to update the tokens inside it.</p>
<p>Set the connection timeout. This should be in go time format which looks like <code>5s</code> for 5 seconds, <code>10m</code> for 10 minutes, or <code>3h30m</code>.</p>
<p>The connection timeout is the amount of time rclone will wait for a connection to go through to a remote object storage system. It is <code>1m</code> by default.</p>
<p>When using <code>sync</code>, <code>copy</code> or <code>move</code> DIR is checked in addition to the destination for files. If a file identical to the source is found that file is server-side copied from DIR to the destination. This is useful for incremental backup.</p>
<p>The remote in use must support server-side copy and you must use the same remote as the destination of the sync. The compare directory must not overlap the destination directory.</p>
<p>Mode to run dedupe command in. One of <code>interactive</code>, <code>skip</code>, <code>first</code>, <code>newest</code>, <code>oldest</code>, <code>rename</code>. The default is <code>interactive</code>. See the dedupe command for more information as to what these options mean.</p>
<p>See the overview <ahref="https://rclone.org/overview/#features">features</a> and <ahref="https://rclone.org/overview/#optional-features">optional features</a> to get an idea of which feature does what.</p>
<p>This flag can be useful for debugging and in exceptional circumstances (e.g. Google Drive limiting the total volume of Server Side Copies to 100GB/day).</p>
<p>Specify a DSCP value or name to use in connections. This could help QoS system to identify traffic class. BE, EF, DF, LE, CSx and AFxx are allowed.</p>
<p>See the description of <ahref="https://en.wikipedia.org/wiki/Differentiated_services">differentiated services</a> to get an idea of this field. Setting this to 1 (LE) to identify the flow to SCAVENGER class can avoid occupying too much bandwidth in a network with DiffServ support (<ahref="https://tools.ietf.org/html/rfc8622">RFC 8622</a>).</p>
<p>For example, if you configured QoS on router to handle LE properly. Running:</p>
<pre><code>rclone copy --dscp LE from:/from to:/to</code></pre>
<p>would make the priority lower than usual internet flows.</p>
<p>Do a trial run with no permanent changes. Use this to see what rclone would do without actually doing it. Useful when setting up the <code>sync</code> command which deletes files in the destination.</p>
<p>This specifies the amount of time to wait for a server's first response headers after fully writing the request headers if the request has an "Expect: 100-continue" header. Not all backends support using this.</p>
<p>Zero means no timeout and causes the body to be sent immediately, without waiting for the server to approve. This time does not include the time to send the request header.</p>
<p>By default, rclone will exit with return code 0 if there were no errors.</p>
<p>This option allows rclone to return exit code 9 if no files were transferred between the source and destination. This allows using rclone in scripts, and triggering follow-on actions if data was copied, or skipping if not.</p>
<p>NB: Enabling this option turns a usually non-fatal error into a potentially fatal one - please check and adjust your scripts accordingly!</p>
<p>When using rclone via the API rclone caches created remotes for 5 minutes by default in the "fs cache". This means that if you do repeated actions on the same remote then rclone won't have to build it again from scratch, which makes it more efficient.</p>
<p>This flag sets the time that the remotes are cached for. If you set it to <code>0</code> (or negative) then rclone won't cache the remotes at all.</p>
<p>Note that if you use some flags, eg <code>--backup-dir</code> and if this is set to <code>0</code> rclone may build two remotes (one for the source or destination and one for the <code>--backup-dir</code> where it may have only built one before.</p>
<p>This controls how often rclone checks for cached remotes to expire. See the <code>--fs-cache-expire-duration</code> documentation above for more info. The default is 60s, set to 0 to disable expiry.</p>
<p>Add an HTTP header for all transactions. The flag can be repeated to add multiple headers.</p>
<p>If you want to add headers only for uploads use <code>--header-upload</code> and if you want to add headers only for downloads use <code>--header-download</code>.</p>
<p>This flag is supported for all HTTP based backends even those not supported by <code>--header-upload</code> and <code>--header-download</code> so may be used as a workaround for those with care.</p>
<pre><code>rclone ls remote:test --header "X-Rclone: Foo" --header "X-LetMeIn: Yes"</code></pre>
<p>Using this option will cause rclone to ignore the case of the files when synchronizing so files will not be copied/synced when the existing filenames are the same, even if the casing is different.</p>
<p>Normally rclone will check that the checksums of transferred files match, and give an error "corrupted on transfer" if they don't.</p>
<p>You can use this option to skip that check. You should only use it if you have had the "corrupted on transfer" error message and you are sure you might want to transfer potentially corrupted data.</p>
<p>While this isn't a generally recommended option, it can be useful in cases where your files change due to encryption. However, it cannot correct partial transfers in case a transfer was interrupted.</p>
<p>Normally rclone will look at modification time and size of files to see if they are equal. If you set this flag then rclone will check only the modification time. If <code>--checksum</code> is set then it only checks the checksum.</p>
<p>It will also cause rclone to skip verifying the sizes are the same after transfer.</p>
<p>This can be useful for transferring files to and from OneDrive which occasionally misreports the size of image files (see <ahref="https://github.com/rclone/rclone/issues/399">#399</a> for more info).</p>
<p>Using this option will cause rclone to unconditionally upload all files regardless of the state of files on the destination.</p>
<p>Normally rclone would skip any files that have the same modification time and are the same size (or have the same checksum if using <code>--checksum</code>).</p>
<p>Treat source and destination files as immutable and disallow modification.</p>
<p>With this option set, files will be created and deleted as requested, but existing files will never be updated. If an existing file does not match between the source and destination, rclone will give the error <code>Source and destination exist but do not match: immutable file modified</code>.</p>
<p>Note that only commands which transfer files (e.g. <code>sync</code>, <code>copy</code>, <code>move</code>) are affected by this behavior, and only modification is disallowed. Files may still be deleted explicitly (e.g. <code>delete</code>, <code>purge</code>) or implicitly (e.g. <code>sync</code>, <code>move</code>). Use <code>copy --immutable</code> if it is desired to avoid deletion as well as modification.</p>
<p>This can be useful as an additional layer of protection for immutable or append-only data sets (notably backup archives), where modification implies corruption and should not be propagated.</p>
<p>This flag can be used to tell rclone that you wish a manual confirmation before destructive operations.</p>
<p>It is <strong>recommended</strong> that you use this flag while learning rclone especially with <code>rclone sync</code>.</p>
<p>For example</p>
<pre><code>$ rclone delete -i /tmp/dir
rclone: delete "important-file.txt"?
y) Yes, this is OK (default)
n) No, skip this
s) Skip all delete operations with no more questions
!) Do all delete operations with no more questions
q) Exit rclone now.
y/n/s/!/q> n</code></pre>
<p>The options mean</p>
<ul>
<li><code>y</code>: <strong>Yes</strong>, this operation should go ahead. You can also press Return for this to happen. You'll be asked every time unless you choose <code>s</code> or <code>!</code>.</li>
<li><code>n</code>: <strong>No</strong>, do not do this operation. You'll be asked every time unless you choose <code>s</code> or <code>!</code>.</li>
<li><code>s</code>: <strong>Skip</strong> all the following operations of this type with no more questions. This takes effect until rclone exits. If there are any different kind of operations you'll be prompted for them.</li>
<li><code>!</code>: <strong>Do all</strong> the following operations with no more questions. Useful if you've decided that you don't mind rclone doing that kind of operation. This takes effect until rclone exits . If there are any different kind of operations you'll be prompted for them.</li>
<li><code>q</code>: <strong>Quit</strong> rclone now, just in case!</li>
</ul>
<h3id="leave-root">--leave-root</h3>
<p>During rmdirs it will not remove root directory, even if it's empty.</p>
<h3id="log-filefile">--log-file=FILE</h3>
<p>Log all of rclone's output to FILE. This is not active by default. This can be useful for tracking down problems with syncs in combination with the <code>-v</code> flag. See the <ahref="#logging">Logging section</a> for more info.</p>
<p>Note that if you are using the <code>logrotate</code> program to manage rclone's logs, then you should use the <code>copytruncate</code> option as rclone doesn't have a signal to rotate logs.</p>
<h3id="log-format-list">--log-format LIST</h3>
<p>Comma separated list of log format options. <code>date</code>, <code>time</code>, <code>microseconds</code>, <code>longfile</code>, <code>shortfile</code>, <code>UTC</code>. The default is "<code>date</code>,<code>time</code>".</p>
<p><code>DEBUG</code> is equivalent to <code>-vv</code>. It outputs lots of debug info - useful for bug reports and really finding out what rclone is doing.</p>
<p><code>INFO</code> is equivalent to <code>-v</code>. It outputs information about each transfer and prints stats once a minute by default.</p>
<p><code>NOTICE</code> is the default log level if no logging flags are supplied. It outputs very little when things are working normally. It outputs warnings and significant events.</p>
<p>This controls the number of low level retries rclone does.</p>
<p>A low level retry is used to retry a failing operation - typically one HTTP request. This might be uploading a chunk of a big file for example. You will see low level retries in the log with the <code>-v</code> flag.</p>
<p>This shouldn't need to be changed from the default in normal operations. However, if you get a lot of low level retries you may wish to reduce the value so rclone moves on to a high level retry (see the <code>--retries</code> flag) quicker.</p>
<p>This is the maximum allowable backlog of files in a sync/copy/move queued for being checked or transferred.</p>
<p>This can be set arbitrarily large. It will only use memory when the queue is in use. Note that it will use in the order of N kB of memory when the backlog is in use.</p>
<p>Setting this large allows rclone to calculate how many files are pending more accurately, give a more accurate estimated finish time and make <code>--order-by</code> work more accurately.</p>
<p>This tells rclone not to delete more than N files. If that limit is exceeded then a fatal error will be generated and rclone will stop the operation in progress.</p>
<p>This modifies the recursion depth for all the commands except purge.</p>
<p>So if you do <code>rclone --max-depth 1 ls remote:path</code> you will see only the files in the top level directory. Using <code>--max-depth 2</code> means you will see all the files in first two directory levels and so on.</p>
<p>For historical reasons the <code>lsd</code> command defaults to using a <code>--max-depth</code> of 1 - you can override this with the command line flag.</p>
<p>You can use this command to disable recursion (with <code>--max-depth 1</code>).</p>
<p>Note that if you use this with <code>sync</code> and <code>--delete-excluded</code> the files not recursed through are considered excluded and will be deleted on the destination. Test first with <code>--dry-run</code> if you are not sure what will happen.</p>
<p>When checking whether a file has been modified, this is the maximum allowed time difference that a file can have and still be considered equivalent.</p>
<p>The default is <code>1ns</code> unless this is overridden by a remote. For example OS X only stores modification times to the nearest second so if you are reading and writing to an OS X filing system this will be <code>1s</code> by default.</p>
<p>This command line flag allows you to override that computed default.</p>
<p>Rclone preallocates the file (using <code>fallocate(FALLOC_FL_KEEP_SIZE)</code> on unix or <code>NTSetInformationFile</code> on Windows both of which takes no time) then each thread writes directly into the file at the correct place. This means that rclone won't create fragmented or sparse files and there won't be any assembly time at the end of the transfer.</p>
<p>Use <code>-vv</code> if you wish to see info about the threads.</p>
<p>This will work with the <code>sync</code>/<code>copy</code>/<code>move</code> commands and friends <code>copyto</code>/<code>moveto</code>. Multi thread downloads will be used with <code>rclone mount</code> and <code>rclone serve</code> if <code>--vfs-cache-mode</code> is set to <code>writes</code> or above.</p>
<p><strong>NB</strong> that this <strong>only</strong> works for a local destination but will work with any source.</p>
<p><strong>NB</strong> that multi thread copies are disabled for local to local copies as they are faster without unless <code>--multi-thread-streams</code> is set explicitly.</p>
<p><strong>NB</strong> on Windows using multi-thread downloads will cause the resulting files to be <ahref="https://en.wikipedia.org/wiki/Sparse_file">sparse</a>. Use <code>--local-no-sparse</code> to disable sparse files (which may cause long delays at the start of downloads) or disable multi-thread downloads with <code>--multi-thread-streams 0</code></p>
<p>When using multi thread downloads (see above <code>--multi-thread-cutoff</code>) this sets the maximum number of streams to use. Set to <code>0</code> to disable multi thread downloads (Default 4).</p>
<p>Exactly how many streams rclone uses for the download depends on the size of the file. To calculate the number of download streams Rclone divides the size of the file by the <code>--multi-thread-cutoff</code> and rounds up, up to the maximum set with <code>--multi-thread-streams</code>.</p>
<p>So if <code>--multi-thread-cutoff 250MB</code> and <code>--multi-thread-streams 4</code> are in effect (the defaults):</p>
<p>The <code>--no-check-dest</code> can be used with <code>move</code> or <code>copy</code> and it causes rclone not to check the destination at all when copying files.</p>
<p>This means that:</p>
<ul>
<li>the destination is not listed minimising the API calls</li>
<p>Don't set <code>Accept-Encoding: gzip</code>. This means that rclone won't ask the server for compressed files automatically. Useful if you've set the server to return files with <code>Content-Encoding: gzip</code> but you uploaded compressed files.</p>
<p>The <code>--no-traverse</code> flag controls whether the destination file system is traversed when using the <code>copy</code> or <code>move</code> commands. <code>--no-traverse</code> is not compatible with <code>sync</code> and will be ignored if you supply it with <code>sync</code>.</p>
<p>If you are only copying a small number of files (or are filtering most of the files) and/or have a large number of files on the destination then <code>--no-traverse</code> will stop rclone listing the destination and save time.</p>
<p>However, if you are copying a large number of files, especially if you are doing a copy where lots of the files under consideration haven't changed and won't need copying then you shouldn't use <code>--no-traverse</code>.</p>
<p>Sometimes, an operating system will store filenames containing unicode parts in their decomposed form (particularly macOS). Some cloud storage systems will then recompose the unicode, resulting in duplicate files if the data is ever copied back to a local filesystem.</p>
<p>Using this flag will disable that functionality, treating each unicode character as unique. For example, by default é and é will be normalized into the same character. With <code>--no-unicode-normalization</code> they will be treated as unique characters.</p>
<p>The <code>--order-by</code> flag controls the order in which files in the backlog are processed in <code>rclone sync</code>, <code>rclone copy</code> and <code>rclone move</code>.</p>
<p>The order by string is constructed like this. The first part describes what aspect is being measured:</p>
<ul>
<li><code>size</code> - order by the size of the files</li>
<li><code>name</code> - order by the full path of the files</li>
<li><code>modtime</code> - order by the modification date of the files</li>
</ul>
<p>This can have a modifier appended with a comma:</p>
<ul>
<li><code>ascending</code> or <code>asc</code> - order so that the smallest (or oldest) is processed first</li>
<li><code>descending</code> or <code>desc</code> - order so that the largest (or newest) is processed first</li>
<p>If the modifier is <code>mixed</code> then it can have an optional percentage (which defaults to <code>50</code>), e.g. <code>size,mixed,25</code> which means that 25% of the threads should be taking the smallest items and 75% the largest. The threads which take the smallest first will always take the smallest first and likewise the largest first threads. The <code>mixed</code> mode can be useful to minimise the transfer time when you are transferring a mixture of large and small files - the large files are guaranteed upload threads and bandwidth and the small files will be processed continuously.</p>
<p>If no modifier is supplied then the order is <code>ascending</code>.</p>
<p>For example</p>
<ul>
<li><code>--order-by size,desc</code> - send the largest files first</li>
<li><code>--order-by modtime,ascending</code> - send the oldest files first</li>
<li><code>--order-by name</code> - send the files with alphabetically by path first</li>
</ul>
<p>If the <code>--order-by</code> flag is not supplied or it is supplied with an empty string then the default ordering will be used which is as scanned. With <code>--checkers 1</code> this is mostly alphabetical, however with the default <code>--checkers 8</code> it is somewhat random.</p>
<h4id="limitations-1">Limitations</h4>
<p>The <code>--order-by</code> flag does not do a separate pass over the data. This means that it may transfer some files out of the order specified if</p>
<ul>
<li>there are no files in the backlog or the source has not been fully scanned yet</li>
<p>Rclone will do its best to transfer the best file it has so in practice this should not cause a problem. Think of <code>--order-by</code> as being more of a best efforts flag rather than a perfect ordering.</p>
<p>If you want perfect ordering then you will need to specify <ahref="#check-first">--check-first</a> which will find all the files which need transferring first before transferring any.</p>
<p>This flag supplies a program which should supply the config password when run. This is an alternative to rclone prompting for the password or setting the <code>RCLONE_CONFIG_PASS</code> variable.</p>
<p>The argument to this should be a command with a space separated list of arguments. If one of the arguments has a space in then enclose it in <code>"</code>, if you want a literal <code>"</code> in an argument then enclose the argument in <code>"</code> and double the <code>"</code>. See <ahref="https://godoc.org/encoding/csv">CSV encoding</a> for more info.</p>
<p>Eg</p>
<pre><code>--password-command echo hello
--password-command echo "hello with space"
--password-command echo "hello with ""quotes"" and space"</code></pre>
<p>See the <ahref="#configuration-encryption">Configuration Encryption</a> for more info.</p>
<p>See a <ahref="https://github.com/rclone/rclone/wiki/Windows-Powershell-use-rclone-password-command-for-Config-file-password">Windows PowerShell example on the Wiki</a>.</p>
<p>This flag makes rclone update the stats in a static block in the terminal providing a realtime overview of the transfer.</p>
<p>Any log messages will scroll above the static block. Log messages will push the static block down to the bottom of the terminal where it will stay.</p>
<p>Normally this is updated every 500mS but this period can be overridden with the <code>--stats</code> flag.</p>
<p>This can be used with the <code>--stats-one-line</code> flag for a simpler display.</p>
<p>Note: On Windows until <ahref="https://github.com/Azure/go-ansiterm/issues/26">this bug</a> is fixed all non-ASCII characters will be replaced with <code>.</code> when <code>--progress</code> is in use.</p>
<p>This flag will limit rclone's output to error messages only.</p>
<h3id="refresh-times">--refresh-times</h3>
<p>The <code>--refresh-times</code> flag can be used to update modification times of existing files when they are out of sync on backends which don't support hashes.</p>
<p>This is useful if you uploaded files with the incorrect timestamps and you now wish to correct them.</p>
<p>This can be used any of the sync commands <code>sync</code>, <code>copy</code> or <code>move</code>.</p>
<p>To use this flag you will need to be doing a modification time sync (so not using <code>--size-only</code> or <code>--checksum</code>). The flag will have no effect when using <code>--size-only</code> or <code>--checksum</code>.</p>
<p>If this flag is used when rclone comes to upload a file it will check to see if there is an existing file on the destination. If this file matches the source with size (and checksum if available) but has a differing timestamp then instead of re-uploading it, rclone will update the timestamp on the destination file. If the checksum does not match rclone will upload the new file. If the checksum is absent (e.g. on a <code>crypt</code> backend) then rclone will update the timestamp.</p>
<p>Note that some remotes can't set the modification time without re-uploading the file so this flag is less useful on them.</p>
<p>Normally if you are doing a modification time sync rclone will update modification times without <code>--refresh-times</code> provided that the remote supports checksums <strong>and</strong> the checksums match on the file. However if the checksums are absent then rclone will upload the file rather than setting the timestamp as this is the safe behaviour.</p>
<p>Normally rclone will look at modification time and size of files to see if they are equal. If you set this flag then rclone will check only the size.</p>
<p>This can be useful transferring files from Dropbox which have been modified by the desktop sync client which doesn't set checksums of modification times in the same way as rclone.</p>
<p>Commands which transfer data (<code>sync</code>, <code>copy</code>, <code>copyto</code>, <code>move</code>, <code>moveto</code>) will print data transfer stats at regular intervals to show their progress.</p>
<p>If you set the stats interval then all commands can show stats. This can be useful when running other commands, <code>check</code> or <code>mount</code> for example.</p>
<p>Stats are logged at <code>INFO</code> level by default which means they won't show at default log level <code>NOTICE</code>. Use <code>--stats-log-level NOTICE</code> or <code>-v</code> to make them show. See the <ahref="#logging">Logging section</a> for more info on log levels.</p>
<p>By default, the <code>--stats</code> output will truncate file names and paths longer than 40 characters. This is equivalent to providing <code>--stats-file-name-length 40</code>. Use <code>--stats-file-name-length 0</code> to disable any truncation of file names printed by stats.</p>
<p>Log level to show <code>--stats</code> output at. This can be <code>DEBUG</code>, <code>INFO</code>, <code>NOTICE</code>, or <code>ERROR</code>. The default is <code>INFO</code>. This means at the default level of logging which is <code>NOTICE</code> the stats won't show - if you want them to then use <code>--stats-log-level NOTICE</code>. See the <ahref="#logging">Logging section</a> for more info on log levels.</p>
<p>When this is specified, rclone enables the single-line stats and prepends the display with a date string. The default is <code>2006/01/02 15:04:05 -</code></p>
<p>When this is specified, rclone enables the single-line stats and prepends the display with a user-supplied date string. The date string MUST be enclosed in quotes. Follow <ahref="https://golang.org/pkg/time/#Time.Format">golang specs</a> for date formatting syntax.</p>
<p>When using <code>sync</code>, <code>copy</code> or <code>move</code> any files which would have been overwritten or deleted will have the suffix added to them. If there is a file with the same path (after the suffix has been added), then it will be overwritten.</p>
<p>This is for use with files to add the suffix in the current directory or with <code>--backup-dir</code>. See <code>--backup-dir</code> for more info.</p>
<p>will copy <code>/path/to/local</code> to <code>remote:current</code>, but for any files which would have been updated or deleted have .bak added.</p>
<p>If using <code>rclone sync</code> with <code>--suffix</code> and without <code>--backup-dir</code> then it is recommended to put a filter rule in excluding the suffix otherwise the <code>sync</code> will delete the backup files.</p>
<p>When using <code>--suffix</code>, setting this causes rclone put the SUFFIX before the extension of the files that it backs up rather than after.</p>
<p>So let's say we had <code>--suffix -2019-01-01</code>, without the flag <code>file.txt</code> would be backed up to <code>file.txt-2019-01-01</code> and with the flag it would be backed up to <code>file-2019-01-01.txt</code>. This can be helpful to make sure the suffixed files can still be opened.</p>
<p>If using <code>--syslog</code> this sets the syslog facility (e.g. <code>KERN</code>, <code>USER</code>). See <code>man syslog</code> for a list of possible facilities. The default facility is <code>DAEMON</code>.</p>
<p>Limit transactions per second to this number. Default is 0 which is used to mean unlimited transactions per second.</p>
<p>A transaction is roughly defined as an API call; its exact meaning will depend on the backend. For HTTP based backends it is an HTTP PUT/GET/POST/etc and its response. For FTP/SFTP it is a round trip transaction over TCP.</p>
<p>For example to limit rclone to 10 transactions per second use <code>--tpslimit 10</code>, or to 1 transaction every 2 seconds use <code>--tpslimit 0.5</code>.</p>
<p>Use this when the number of transactions per second from rclone is causing a problem with the cloud storage provider (e.g. getting you banned or rate limited).</p>
<p>This limit applies to all HTTP based backends and to the FTP and SFTP backends. It does not apply to the local backend or the Tardigrade backend.</p>
<p>Normally <code>--tpslimit</code> will do exactly the number of transaction per second specified. However if you supply <code>--tps-burst</code> then rclone can save up some transactions from when it was idle giving a burst of up to the parameter supplied.</p>
<p>For example if you provide <code>--tpslimit-burst 10</code> then if rclone has been idle for more than 10*<code>--tpslimit</code> then it can do 10 transactions very quickly before they are limited again.</p>
<p>This may be used to increase performance of <code>--tpslimit</code> without changing the long term average number of transactions per second.</p>
<p>By default, rclone doesn't keep track of renamed files, so if you rename a file locally then sync it to a remote, rclone will delete the old file on the remote and upload a new copy.</p>
<p>If you use this flag, and the remote supports server-side copy or server-side move, and the source and destination have a compatible hash, then this will track renames during <code>sync</code> operations and perform renaming server-side.</p>
<p>If the destination does not support server-side copy or move, rclone will fall back to the default behaviour and log an error level message to the console.</p>
<p>Encrypted destinations are not currently supported by <code>--track-renames</code> if <code>--track-renames-strategy</code> includes <code>hash</code>.</p>
<p>Note that <code>--track-renames</code> is incompatible with <code>--no-traverse</code> and that it uses extra memory to keep track of all the rename candidates.</p>
<p>Note also that <code>--track-renames</code> is incompatible with <code>--delete-before</code> and will select <code>--delete-after</code> instead of <code>--delete-during</code>.</p>
<p>This option changes the matching criteria for <code>--track-renames</code>.</p>
<p>The matching is controlled by a comma separated selection of these tokens:</p>
<ul>
<li><code>modtime</code> - the modification time of the file - not supported on all backends</li>
<li><code>hash</code> - the hash of the file contents - not supported on all backends</li>
<li><code>leaf</code> - the name of the file not including its directory name</li>
<li><code>size</code> - the size of the file (this is always enabled)</li>
</ul>
<p>So using <code>--track-renames-strategy modtime,leaf</code> would match files based on modification time, the leaf of the file name and the size only.</p>
<p>Using <code>--track-renames-strategy modtime</code> or <code>leaf</code> can enable <code>--track-renames</code> support for encrypted destinations.</p>
<p>If nothing is specified, the default option is matching by <code>hash</code>es.</p>
<p>Note that the <code>hash</code> strategy is not supported with encrypted destinations.</p>
<p>Specifying the value <code>--delete-before</code> will delete all files present on the destination, but not on the source <em>before</em> starting the transfer of any new or updated files. This uses two passes through the file systems, one for the deletions and one for the copies.</p>
<p>Specifying <code>--delete-during</code> will delete files while checking and uploading files. This is the fastest option and uses the least memory.</p>
<p>Specifying <code>--delete-after</code> (the default value) will delay deletion of files until all new/updated files have been successfully transferred. The files to be deleted are collected in the copy pass then deleted after the copy pass has completed successfully. The files to be deleted are held in memory so this mode may use more memory. This is the safest mode as it will only delete files if there have been no errors subsequent to that. If there have been errors before the deletions start then you will get the message <code>not deleting files as there were IO errors</code>.</p>
<p>When doing anything which involves a directory listing (e.g. <code>sync</code>, <code>copy</code>, <code>ls</code> - in fact nearly every command), rclone normally lists a directory and processes it before using more directory lists to process any subdirectories. This can be parallelised and works very quickly using the least amount of memory.</p>
<p>However, some remotes have a way of listing all files beneath a directory in one (or a small number) of transactions. These tend to be the bucket based remotes (e.g. S3, B2, GCS, Swift, Hubic).</p>
<p>If you use the <code>--fast-list</code> flag then rclone will use this method for listing directories. This will have the following consequences for the listing:</p>
<ul>
<li>It <strong>will</strong> use fewer transactions (important if you pay for them)</li>
<li>It <strong>will</strong> use more memory. Rclone has to load the whole listing into memory.</li>
<li>It <em>may</em> be faster because it uses fewer transactions</li>
<p>If you pay for transactions and can fit your entire sync listing into memory then <code>--fast-list</code> is recommended. If you have a very big sync to do then don't use <code>--fast-list</code> otherwise you will run out of memory.</p>
<p>If you use <code>--fast-list</code> on a remote which doesn't support it, then rclone will just ignore it.</p>
<p>The number of file transfers to run in parallel. It can sometimes be useful to set this to a smaller number if the remote is giving a lot of timeouts or bigger if you have lots of bandwidth and a fast remote.</p>
<p>The default is to run 4 file transfers in parallel.</p>
<p>This can be useful when transferring to a remote which doesn't support mod times directly (or when using <code>--use-server-modtime</code> to avoid extra API calls) as it is more accurate than a <code>--size-only</code> check and faster than using <code>--checksum</code>.</p>
<p>If an existing destination file has a modification time equal (within the computed modify window precision) to the source file's, it will be updated if the sizes are different. If <code>--checksum</code> is set then rclone will update the destination if the checksums differ too.</p>
<p>On remotes which don't support mod time directly (or when using <code>--use-server-modtime</code>) the time checked will be the uploaded time. This means that if uploading to one of these remotes, rclone will skip any files which exist on the destination and have an uploaded time that is newer than the modification time of the source file.</p>
<p>If this flag is set then rclone will use anonymous memory allocated by mmap on Unix based platforms and VirtualAlloc on Windows for its transfer buffers (size controlled by <code>--buffer-size</code>). Memory allocated like this does not go on the Go heap and can be returned to the OS immediately when it is finished with.</p>
<p>If this flag is not set then rclone will allocate and free the buffers using the Go memory allocator which may use more memory as memory pages are returned less aggressively to the OS.</p>
<p>It is possible this does not work well on all platforms so it is disabled by default; in the future it may be enabled by default.</p>
<p>Some object-store backends (e.g, Swift, S3) do not preserve file modification times (modtime). On these backends, rclone stores the original modtime as additional metadata on the object. By default it will make an API call to retrieve the metadata when the modtime is needed by an operation.</p>
<p>Use this flag to disable the extra API call and rely instead on the server's modified time. In cases such as a local to remote sync using <code>--update</code>, knowing the local file is newer than the time it was last uploaded to the remote is sufficient. In those cases, this flag can speed up the process and reduce the number of API calls necessary.</p>
<p>Using this flag on a sync operation without also using <code>--update</code> would cause all files modified at any time other than the last upload time to be uploaded again, which is probably not what you want.</p>
<p>With <code>-v</code> rclone will tell you about each file that is transferred and a small number of significant events.</p>
<p>With <code>-vv</code> rclone will become very verbose telling you about every file it considers and transfers. Please send bug reports with a log with this setting.</p>
<p>The outgoing SSL/TLS connections rclone makes can be controlled with these options. For example this can be very useful with the HTTP or WebDAV backends. Rclone HTTP servers have their own set of configuration for SSL/TLS which you can find in their documentation.</p>
<p><code>--no-check-certificate</code> controls whether a client verifies the server's certificate chain and host name. If <code>--no-check-certificate</code> is true, TLS accepts any certificate presented by the server and any host name in that certificate. In this mode, TLS is susceptible to man-in-the-middle attacks.</p>
<p>Your configuration file contains information for logging in to your cloud services. This means that you should keep your <code>rclone.conf</code> file in a secure location.</p>
<p>If you are in an environment where that isn't possible, you can add a password to your configuration. This means that you will have to supply the password every time you start rclone.</p>
<p>Your configuration is now encrypted, and every time you start rclone you will have to supply the password. See below for details. In the same menu, you can change the password or completely remove encryption from your configuration.</p>
<p>rclone uses <ahref="https://godoc.org/golang.org/x/crypto/nacl/secretbox">nacl secretbox</a> which in turn uses XSalsa20 and Poly1305 to encrypt and authenticate your configuration with secret-key cryptography. The password is SHA-256 hashed, which produces the key for secretbox. The hashed password is not stored.</p>
<p>While this provides very good security, we do not recommend storing your encrypted rclone configuration in public if it contains sensitive information, maybe except if you use a very strong password.</p>
<p>If it is safe in your environment, you can set the <code>RCLONE_CONFIG_PASS</code> environment variable to contain your password, in which case it will be used for decrypting the configuration.</p>
<p>Then source the file when you want to use it. From the shell you would do <code>source set-rclone-password</code>. It will then ask you for the password and set it in the environment variable.</p>
<p>An alternate means of supplying the password is to provide a script which will retrieve the password and print on standard output. This script should have a fully specified path name and not rely on any environment variables. The script is supplied either via <code>--password-command="..."</code> command line argument or via the <code>RCLONE_PASSWORD_COMMAND</code> environment variable.</p>
<p>One useful example of this is using the <code>passwordstore</code> application to retrieve the password:</p>
<p>If the <code>passwordstore</code> password manager holds the password for the rclone configuration, using the script method means the password is primarily protected by the <code>passwordstore</code> system, and is never embedded in the clear in scripts, nor available for examination using the standard commands available. It is quite possible with long running rclone sessions for copies of passwords to be innocently captured in log files or terminal scroll buffers, etc. Using the script method of supplying the password enhances the security of the config password considerably.</p>
<p>If you are running rclone inside a script, unless you are using the <code>--password-command</code> method, you might want to disable password prompts. To do that, pass the parameter <code>--ask-password=false</code> to rclone. This will make rclone fail instead of asking for a password if <code>RCLONE_CONFIG_PASS</code> doesn't contain a valid password, and <code>--password-command</code> has not been supplied.</p>
<p>Some rclone commands, such as <code>genautocomplete</code>, do not require configuration. Nevertheless, rclone will read any configuration file found according to the rules described <ahref="https://rclone.org/docs/#config-config-file">above</a>. If an encrypted configuration file is found, this means you will be prompted for password (unless using <code>--password-command</code>). To avoid this, you can bypass the loading of the configuration file by overriding the location with an empty string <code>""</code> or the special value <code>/notfound</code>, or the os null device represented by value <code>NUL</code> on Windows and <code>/dev/null</code> on Unix systems (before rclone version 1.55 only this null device alternative was supported). E.g. <code>rclone --config="" genautocomplete bash</code>.</p>
<p>These options are useful when developing or debugging rclone. There are also some more remote specific options which aren't documented here which are used for testing. These start with remote name e.g. <code>--drive-test-option</code> - see the docs for the remote in question.</p>
<p>The <code>--dump</code> flag takes a comma separated list of flags to dump info about.</p>
<p>Note that some headers including <code>Accept-Encoding</code> as shown may not be correct in the request and the response may not show <code>Content-Encoding</code> if the go standard libraries auto gzip encoding was in effect. In this case the body of the request will be gunzipped before showing it.</p>
<p>Dump HTTP headers with <code>Authorization:</code> lines removed. May still contain sensitive info. Can be very verbose. Useful for debugging only.</p>
<p>Use <code>--dump auth</code> if you do want the <code>Authorization:</code> headers.</p>
<p>Dump HTTP headers - will contain sensitive info such as <code>Authorization:</code> headers - use <code>--dump headers</code> to dump without <code>Authorization:</code> headers. Can be very verbose. Useful for debugging only.</p>
<p>This dumps a list of the open files at the end of the command. It uses the <code>lsof</code> command to do that so you'll need that installed to use it.</p>
<p>By default, rclone logs to standard error. This means you can redirect standard error and still see the normal output of rclone commands (e.g. <code>rclone ls</code>).</p>
<p>If you use the <code>-v</code> flag, rclone will produce <code>Error</code>, <code>Notice</code> and <code>Info</code> messages.</p>
<p>If you use the <code>-vv</code> flag, rclone will produce <code>Error</code>, <code>Notice</code>, <code>Info</code> and <code>Debug</code> messages.</p>
<p>You can also control the log levels with the <code>--log-level</code> flag.</p>
<p>If you use the <code>--log-file=FILE</code> option, rclone will redirect <code>Error</code>, <code>Info</code> and <code>Debug</code> messages along with standard error to FILE.</p>
<p>Rclone prefixes all log messages with their level in capitals, e.g. INFO which makes it easy to grep the log file for different kinds of information.</p>
<p>If any errors occur during the command execution, rclone will exit with a non-zero exit code. This allows scripts to detect when rclone operations have failed.</p>
<p>During the startup phase, rclone will exit immediately if an error is detected in the configuration. There will always be a log message immediately before exiting.</p>
<p>When rclone is running it will accumulate errors as it goes along, and only exit with a non-zero exit code if (after retries) there were still failed transfers. For every error counted there will be a high priority log message (visible with <code>-q</code>) showing the message and which file caused the problem. A high priority message is also shown when starting a retry so the user can see that any previous error messages may not be valid after the retry. If rclone has done a retry it will log a high priority message if the retry was successful.</p>
<p>To find the name of the environment variable, first, take the long option name, strip the leading <code>--</code>, change <code>-</code> to <code>_</code>, make upper case and prepend <code>RCLONE_</code>.</p>
<p>For example, to always set <code>--stats 5s</code>, set the environment variable <code>RCLONE_STATS=5s</code>. If you set stats on the command line this will override the environment variable setting.</p>
<p>You can set defaults for values in the config file on an individual remote basis. The names of the config items are documented in the page for each backend.</p>
<p>To find the name of the environment variable, you need to set, take <code>RCLONE_CONFIG_</code> + name of remote + <code>_</code> + name of config file option and make it all uppercase.</p>
<p>Note also that now rclone has <ahref="#connection-strings">connectionstrings</a>, it is probably easier to use those instead which makes the above example</p>
<p>So if both <code>--drive-use-trash</code> is supplied on the config line and an environment variable <code>RCLONE_DRIVE_USE_TRASH</code> is set, the command line flag will take preference.</p>
<p>For non backend configuration the order is as follows:</p>
<li><code>RCLONE_CONFIG_PASS</code> set to contain your config file password (see <ahref="#configuration-encryption">Configuration Encryption</a> section)</li>
<li><code>HTTP_PROXY</code>, <code>HTTPS_PROXY</code> and <code>NO_PROXY</code> (or the lowercase versions thereof).
<li><code>RCLONE_CONFIG_DIR</code> - rclone <strong>sets</strong> this variable for use in config files and sub processes to point to the directory holding the config file.</li>
<p>If you are trying to set rclone up on a remote or headless box with no browser available on it (e.g. a NAS or a server in a datacenter) then you will need to use an alternative means of configuration. There are two ways of doing it, described below.</p>
<p>Now transfer it to the remote box (scp, cut paste, ftp, sftp, etc.) and place it in the correct place (use <code>rclone config file</code> on the remote box to find out where).</p>
<p>Filter flags determine which files rclone <code>sync</code>, <code>move</code>, <code>ls</code>, <code>lsl</code>, <code>md5sum</code>, <code>sha1sum</code>, <code>size</code>, <code>delete</code>, <code>check</code> and similar commands apply to.</p>
<p>They are specified in terms of path/file name patterns; path/file lists; file age and size, or presence of a file in a directory. Bucket based remotes without the concept of directory apply filters to object key, age and size in an analogous way.</p>
<p>Rclone <code>purge</code> does not obey filters.</p>
<p>To test filters without risk of damage to data, apply them to <code>rclone ls</code>, or with the <code>--dry-run</code> and <code>-vv</code> flags.</p>
<p>Rclone filter patterns can only be used in filter command line options, not in the specification of a remote.</p>
<p>E.g. <code>rclone copy "remote:dir*.jpg" /path/to/dir</code> does not have a filter effect. <code>rclone copy remote:dir /path/to/dir --include "*.jpg"</code> does.</p>
<p><strong>Important</strong> Avoid mixing any two of <code>--include...</code>, <code>--exclude...</code> or <code>--filter...</code> flags in an rclone command. The results may not be what you expect. Instead use a <code>--filter...</code> flag.</p>
<h2id="patterns-for-matching-pathfile-names">Patterns for matching path/file names</h2>
<h3id="pattern-syntax">Pattern syntax</h3>
<p>Rclone matching rules follow a glob style:</p>
<pre><code>`*` matches any sequence of non-separator (`/`) characters
`**` matches any sequence of characters including `/` separators
`?` matches any single non-separator (`/`) character
`[` [ `!` ] { character-range } `]`
character class (must be non-empty)
`{` pattern-list `}`
pattern alternatives
c matches character c (c != `*`, `**`, `?`, `\`, `[`, `{`, `}`)
`\` c matches character c</code></pre>
<p>character-range:</p>
<pre><code>c matches character c (c != `\\`, `-`, `]`)
`\` c matches character c
lo `-` hi matches character c for lo <= c <= hi</code></pre>
<p>character classes (see <ahref="https://golang.org/pkg/regexp/syntax/">Go regular expression reference</a>) include:</p>
<pre><code>Named character classes (e.g. [\d], [^\d], [\D], [^\D])
Perl character classes (e.g. \s, \S, \w, \W)
ASCII character classes (e.g. [[:alnum:]], [[:alpha:]], [[:punct:]], [[:xdigit:]])</code></pre>
<p>If the filter pattern starts with a <code>/</code> then it only matches at the top level of the directory tree, <strong>relative to the root of the remote</strong> (not necessarily the root of the drive). If it does not start with <code>/</code> then it is matched starting at the <strong>end of the path/file name</strong> but it only matches a complete path element - it must match from a <code>/</code> separator or the beginning of the path/file.</p>
<p>The top level of the remote may not be the top level of the drive.</p>
<p>E.g. for a Microsoft Windows local directory structure</p>
<pre><code>F:
├── bkp
├── data
│ ├── excl
│ │ ├── 123.jpg
│ │ └── 456.jpg
│ ├── incl
│ │ └── document.pdf</code></pre>
<p>To copy the contents of folder <code>data</code> into folder <code>bkp</code> excluding the contents of subfolder <code>excl</code>the following command treats <code>F:\data</code> and <code>F:\bkp</code> as top level for filtering.</p>
<p>There can be more than one instance of individual flags.</p>
<p>Rclone internally uses a combined list of all the include and exclude rules. The order in which rules are processed can influence the result of the filter.</p>
<p>All flags of the same type are processed together in the order above, regardless of what order the different types of flags are included on the command line.</p>
<p>Multiple instances of the same flag are processed from left to right according to their position in the command line.</p>
<p>To mix up the order of processing includes and excludes use <code>--filter...</code> flags.</p>
<p>Within <code>--include-from</code>, <code>--exclude-from</code> and <code>--filter-from</code> flags rules are processed from top to bottom of the referenced file.</p>
<p>If there is an <code>--include</code> or <code>--include-from</code> flag specified, rclone implies a <code>- **</code> rule which it adds to the bottom of the internal rule list. Specifying a <code>+</code> rule with a <code>--filter...</code> flag does not imply that rule.</p>
<p>Each path/file name passed through rclone is matched against the combined filter list. At first match to a rule the path/file name is included or excluded and no further filter rules are processed for that path/file.</p>
<p>If rclone does not find a match, after testing against all rules (including the implied rule if appropriate), the path/file name is included.</p>
<p>Any path/file included at that stage is processed by the rclone command.</p>
<p><code>--files-from</code> and <code>--files-from-raw</code> flags over-ride and cannot be combined with other filter options.</p>
<p>To see the internal combined rule list, in regular expression form, for a command add the <code>--dump filters</code> flag. Running an rclone command with <code>--dump filters</code> and <code>-vv</code> flags lists the internal filter elements and shows how they are applied to each source path/file. There is not currently a means provided to pass regular expression filter options into rclone directly though character class filter rules contain character classes. <ahref="https://golang.org/pkg/regexp/syntax/">Go regular expression reference</a></p>
<h3id="how-filter-rules-are-applied-to-directories">How filter rules are applied to directories</h3>
<p>Rclone commands are applied to path/file names not directories. The entire contents of a directory can be matched to a filter by the pattern <code>directory/*</code> or recursively by <code>directory/**</code>.</p>
<p>Directory filter rules are defined with a closing <code>/</code> separator.</p>
<p>E.g. <code>/directory/subdirectory/</code> is an rclone directory filter rule.</p>
<p>Rclone commands can use directory filter rules to determine whether they recurse into subdirectories. This potentially optimises access to a remote by avoiding listing unnecessary directories. Whether optimisation is desirable depends on the specific filter rules and source remote content.</p>
<li><p>A source remote does not support the rclone <code>ListR</code> primitive. local, sftp, Microsoft OneDrive and WebDav do not support <code>ListR</code>. Google Drive and most bucket type storage do. <ahref="https://rclone.org/overview/#optional-features">Full list</a></p></li>
<li><p>On other remotes (those that support <code>ListR</code>), if the rclone command is not naturally recursive, and provided it is not run with the <code>--fast-list</code> flag. <code>ls</code>, <code>lsf -R</code> and <code>size</code> are naturally recursive but <code>sync</code>, <code>copy</code> and <code>move</code> are not.</p></li>
<li><p>Whenever the <code>--disable ListR</code> flag is applied to an rclone command.</p></li>
</ul>
<p>Rclone commands imply directory filter rules from path/file filter rules. To view the directory filter rules rclone has implied for a command specify the <code>--dump filters</code> flag.</p>
<p>E.g. for an include rule</p>
<pre><code>/a/*.jpg</code></pre>
<p>Rclone implies the directory include rule</p>
<pre><code>/a/</code></pre>
<p>Directory filter rules specified in an rclone command can limit the scope of an rclone command but path/file filters still have to be specified.</p>
<p>E.g. <code>rclone ls remote: --include /directory/</code> will not match any files. Because it is an <code>--include</code> option the <code>--exclude **</code> rule is implied, and the <code>/directory/</code> pattern serves only to optimise access to the remote by ignoring everything outside of that directory.</p>
<p>All files in directories <code>dir1</code> or <code>dir2</code> or their subdirectories are completely excluded from the listing. Only files of suffix <code>pdf</code> in the root of <code>remote:</code> or its subdirectories are listed. The <code>- **</code> rule prevents listing of any path/files not previously matched by the rules above.</p>
<p>Option <code>exclude-if-present</code> creates a directory exclude rule based on the presence of a file in a directory and takes precedence over other rclone directory filter rules.</p>
<p>When using pattern list syntax, if a pattern item contains either <code>/</code> or <code>**</code>, then rclone will not able to imply a directory filter rule from this pattern list.</p>
<p>E.g. for an include rule</p>
<pre><code>{dir1/**,dir2/**}</code></pre>
<p>Rclone will match files below directories <code>dir1</code> or <code>dir2</code> only, but will not be able to use this filter to exclude a directory <code>dir3</code> from being traversed.</p>
<p>Directory recursion optimisation may affect performance, but normally not the result. One exception to this is sync operations with option <code>--create-empty-src-dirs</code>, where any traversed empty directories will be created. With the pattern list example <code>{dir1/**,dir2/**}</code> above, this would create an empty directory <code>dir3</code> on destination (when it exists on source). Changing the filter to <code>{dir1,dir2}/**</code>, or splitting it into two include rules <code>--include dir1/** --include dir2/**</code>, will match the same files while also filtering directories, with the result that an empty directory <code>dir3</code> will no longer be created.</p>
<p>Excludes path/file names from an rclone command based on a single exclude rule.</p>
<p>This flag can be repeated. See above for the order filter flags are processed in.</p>
<p><code>--exclude</code> should not be used with <code>--include</code>, <code>--include-from</code>, <code>--filter</code> or <code>--filter-from</code> flags.</p>
<p><code>--exclude</code> has no effect when combined with <code>--files-from</code> or <code>--files-from-raw</code> flags.</p>
<p>E.g. <code>rclone ls remote: --exclude *.bak</code> excludes all .bak files from listing.</p>
<p>E.g. <code>rclone size remote: "--exclude /dir/**"</code> returns the total size of all files on <code>remote:</code> excluding those in root directory <code>dir</code> and sub directories.</p>
<p>E.g. on Microsoft Windows <code>rclone ls remote: --exclude "*\[{JP,KR,HK}\]*"</code> lists the files in <code>remote:</code> with <code>[JP]</code> or <code>[KR]</code> or <code>[HK]</code> in their name. Quotes prevent the shell from interpreting the <code>\</code> characters.<code>\</code> characters escape the <code>[</code> and <code>]</code> so an rclone filter treats them literally rather than as a character-range. The <code>{</code> and <code>}</code> define an rclone pattern list. For other operating systems single quotes are required ie <code>rclone ls remote: --exclude '*\[{JP,KR,HK}\]*'</code></p>
<p><code>rclone ls remote: --exclude-from exclude-file.txt</code> lists the files on <code>remote:</code> except those named <code>file2.jpg</code> or with a suffix <code>.bak</code>. That is equivalent to <code>rclone ls remote: --exclude file2.jpg --exclude "*.bak"</code>.</p>
<p>This flag can be repeated. See above for the order filter flags are processed in.</p>
<p>The <code>--exclude-from</code> flag is useful where multiple exclude filter rules are applied to an rclone command.</p>
<p><code>--exclude-from</code> should not be used with <code>--include</code>, <code>--include-from</code>, <code>--filter</code> or <code>--filter-from</code> flags.</p>
<p><code>--exclude-from</code> has no effect when combined with <code>--files-from</code> or <code>--files-from-raw</code> flags.</p>
<p><code>--exclude-from</code> followed by <code>-</code> reads filter rules from standard input.</p>
<p>Adds a single include rule based on path/file names to an rclone command.</p>
<p>This flag can be repeated. See above for the order filter flags are processed in.</p>
<p><code>--include</code> has no effect when combined with <code>--files-from</code> or <code>--files-from-raw</code> flags.</p>
<p><code>--include</code> implies <code>--exclude **</code> at the end of an rclone internal filter list. Therefore if you mix <code>--include</code> and <code>--include-from</code> flags with <code>--exclude</code>, <code>--exclude-from</code>, <code>--filter</code> or <code>--filter-from</code>, you must use include rules for all the files you want in the include statement. For more flexibility use the <code>--filter-from</code> flag.</p>
<p>E.g. <code>rclone ls remote: --include "*.{png,jpg}"</code> lists the files on <code>remote:</code> with suffix <code>.png</code> and <code>.jpg</code>. All other files are excluded.</p>
<p>E.g. multiple rclone copy commands can be combined with <code>--include</code> and a pattern-list.</p>
<p>E.g. <code>rclone ls remote:/wheat --include "??[^[:punct:]]*"</code> lists the files <code>remote:</code> directory <code>wheat</code> (and subdirectories) whose third character is not punctuation. This example uses an <ahref="https://golang.org/pkg/regexp/syntax/">ASCII character class</a>.</p>
<p><code>rclone ls remote: --include-from include-file.txt</code> lists the files on <code>remote:</code> with name <code>file2.avi</code> or suffix <code>.jpg</code>. That is equivalent to <code>rclone ls remote: --include file2.avi --include "*.jpg"</code>.</p>
<p>This flag can be repeated. See above for the order filter flags are processed in.</p>
<p>The <code>--include-from</code> flag is useful where multiple include filter rules are applied to an rclone command.</p>
<p><code>--include-from</code> implies <code>--exclude **</code> at the end of an rclone internal filter list. Therefore if you mix <code>--include</code> and <code>--include-from</code> flags with <code>--exclude</code>, <code>--exclude-from</code>, <code>--filter</code> or <code>--filter-from</code>, you must use include rules for all the files you want in the include statement. For more flexibility use the <code>--filter-from</code> flag.</p>
<p><code>--exclude-from</code> has no effect when combined with <code>--files-from</code> or <code>--files-from-raw</code> flags.</p>
<p><code>--exclude-from</code> followed by <code>-</code> reads filter rules from standard input.</p>
<p>Specifies path/file names to an rclone command, based on a single include or exclude rule, in <code>+</code> or <code>-</code> format.</p>
<p>This flag can be repeated. See above for the order filter flags are processed in.</p>
<p><code>--filter +</code> differs from <code>--include</code>. In the case of <code>--include</code> rclone implies an <code>--exclude *</code> rule which it adds to the bottom of the internal rule list. <code>--filter...+</code> does not imply that rule.</p>
<p><code>--filter</code> has no effect when combined with <code>--files-from</code> or <code>--files-from-raw</code> flags.</p>
<p><code>--filter</code> should not be used with <code>--include</code>, <code>--include-from</code>, <code>--exclude</code> or <code>--exclude-from</code> flags.</p>
<p>E.g. <code>rclone ls remote: --filter "- *.bak"</code> excludes all <code>.bak</code> files from a list of <code>remote:</code>.</p>
<p>Adds path/file names to an rclone command based on rules in a named file. The file contains a list of remarks and pattern rules. Include rules start with <code>+</code> and exclude rules with <code>-</code>. <code>!</code> clears existing rules. Rules are processed in the order they are defined.</p>
<p>This flag can be repeated. See above for the order filter flags are processed in.</p>
<p>Arrange the order of filter rules with the most restrictive first and work down.</p>
<p><code>rclone ls remote: --filter-from filter-file.txt</code> lists the path/files on <code>remote:</code> including all <code>jpg</code> and <code>png</code> files, excluding any matching <code>secret*.jpg</code> and including <code>file2.avi</code>. It also includes everything in the directory <code>dir</code> at the root of <code>remote</code>, except <code>remote:dir/Trash</code> which it excludes. Everything else is excluded.</p>
<p>E.g. for an alternative <code>filter-file.txt</code>:</p>
<pre><code>- secret*.jpg
+ *.jpg
+ *.png
+ file2.avi
- *</code></pre>
<p>Files <code>file1.jpg</code>, <code>file3.png</code> and <code>file2.avi</code> are listed whilst <code>secret17.jpg</code> and files without the suffix .jpg<code>or</code>.png` are excluded.</p>
<p>E.g. for an alternative <code>filter-file.txt</code>:</p>
<pre><code>+ *.jpg
+ *.gif
!
+ 42.doc
- *</code></pre>
<p>Only file 42.doc is listed. Prior rules are cleared by the <code>!</code>.</p>
<p>Adds path/files to an rclone command from a list in a named file. Rclone processes the path/file names in the order of the list, and no others.</p>
<p>Other filter flags (<code>--include</code>, <code>--include-from</code>, <code>--exclude</code>, <code>--exclude-from</code>, <code>--filter</code> and <code>--filter-from</code>) are ignored when <code>--files-from</code> is used.</p>
<p><code>--files-from</code> expects a list of files as its input. Leading or trailing whitespace is stripped from the input lines. Lines starting with <code>#</code> or <code>;</code> are ignored.</p>
<p>Rclone commands with a <code>--files-from</code> flag traverse the remote, treating the names in <code>--files-from</code> as a set of filters.</p>
<p>If the <code>--no-traverse</code> and <code>--files-from</code> flags are used together an rclone command does not traverse the remote. Instead it addresses each path/file named in the file individually. For each path/file name, that requires typically 1 API call. This can be efficient for a short <code>--files-from</code> list and a remote containing many files.</p>
<p>Rclone commands do not error if any names in the <code>--files-from</code> file are missing from the source remote.</p>
<p>The <code>--files-from</code> flag can be repeated in a single rclone command to read path/file names from more than one file. The files are read from left to right along the command line.</p>
<p>Paths within the <code>--files-from</code> file are interpreted as starting with the root specified in the rclone command. Leading <code>/</code> separators are ignored. See <ahref="#files-from-raw-read-list-of-source-file-names-without-any-processing">--files-from-raw</a> if you need the input to be processed in a raw manner.</p>
<p>E.g. for a file <code>files-from.txt</code>:</p>
<p>E.g. to copy the following files referenced by their absolute paths:</p>
<pre><code>/home/user1/42
/home/user1/dir/ford
/home/user2/prefect</code></pre>
<p>First find a common subdirectory - in this case <code>/home</code> and put the remaining files in <code>files-from.txt</code> with or without leading <code>/</code>, e.g.</p>
<h3id="files-from-raw---read-list-of-source-file-names-without-any-processing"><code>--files-from-raw</code> - Read list of source-file names without any processing</h3>
<p>This flag is the same as <code>--files-from</code> except that input is read in a raw manner. Lines with leading / trailing whitespace, and lines starting with <code>;</code> or <code>#</code> are read without any processing. <ahref="https://rclone.org/commands/rclone_lsf/">rclone lsf</a> has a compatible format that can be used to export file lists from remotes for input to <code>--files-from-raw</code>.</p>
<h3id="ignore-case---make-searches-case-insensitive"><code>--ignore-case</code> - make searches case insensitive</h3>
<p>By default rclone filter patterns are case sensitive. The <code>--ignore-case</code> flag makes all of the filters patterns on the command line case insensitive.</p>
<p>E.g. <code>--include "zaphod.txt"</code> does not match a file <code>Zaphod.txt</code>. With <code>--ignore-case</code> a match is made.</p>
<p>Microsoft Windows expansion is done by the command, not shell, so <code>--include *.jpg</code> does not require quoting.</p>
<p>If the rclone error <code>Command .... needs .... arguments maximum: you provided .... non flag arguments:</code> is encountered, the cause is commonly spaces within the name of a remote or flag value. The fix then is to quote values containing spaces.</p>
<p>Controls the minimum size file within the scope of an rclone command. Default units are <code>kBytes</code> but abbreviations <code>k</code>, <code>M</code>, or <code>G</code> are valid.</p>
<p>E.g. <code>rclone ls remote: --min-size 50k</code> lists files on <code>remote:</code> of 50kByte size or larger.</p>
<p>Controls the maximum size file within the scope of an rclone command. Default units are <code>kBytes</code> but abbreviations <code>k</code>, <code>M</code>, or <code>G</code> are valid.</p>
<p>E.g. <code>rclone ls remote: --max-size 1G</code> lists files on <code>remote:</code> of 1GByte size or smaller.</p>
<p>In conjunction with <code>rclone sync</code>, <code>--delete-excluded</code> deletes any files on the destination which are excluded from the command.</p>
<p>The <code>--exclude-if-present</code> flag controls whether a directory is within the scope of an rclone command based on the presence of a named file within it.</p>
<p>This flag has a priority over other filter flags.</p>
<p>E.g. for the following directory structure:</p>
<p>The command <code>rclone ls --exclude-if-present .ignore dir1</code> does not list <code>dir3</code>, <code>file3</code> or <code>.ignore</code>.</p>
<p><code>--exclude-if-present</code> can only be used once in an rclone command.</p>
<p>Rclone can serve a web based GUI (graphical user interface). This is somewhat experimental at the moment so things may be subject to change.</p>
<p>Run this command in a terminal and rclone will download and then display the GUI in a web browser.</p>
<pre><code>rclone rcd --rc-web-gui</code></pre>
<p>This will produce logs like this and rclone needs to continue to run to serve the GUI:</p>
<pre><code>2019/08/25 11:40:14 NOTICE: A new release for gui is present at https://github.com/rclone/rclone-webui-react/releases/download/v0.0.6/currentbuild.zip
<p>Enable the serving of remote objects via the HTTP interface. This means objects will be accessible at http://127.0.0.1:5572/ by default, so you can browse to http://127.0.0.1:5572/ or http://127.0.0.1:5572/* to see a listing of the remotes. Objects may be requested from remotes using this syntax http://127.0.0.1:5572/[remote:path]/path/to/object</p>
<p>Path to local files to serve on the HTTP server.</p>
<p>If this is set then rclone will serve the files in that directory. It will also open the root in the web browser if specified. This is for implementing browser based GUIs for rclone functions.</p>
<p>If <code>--rc-user</code> or <code>--rc-pass</code> is set then the URL that is opened will have the authorization in the URL in the <code>http://user:pass@localhost/</code> style.</p>
<p>By default rclone will require authorisation to have been set up on the rc interface in order to use any methods which access any rclone remotes. Eg <code>operations/list</code> is denied as it involved creating a remote as is <code>sync/copy</code>.</p>
<p>If this is set then no authorisation will be required on the server to use these methods. The alternative is to use <code>--rc-user</code> and <code>--rc-pass</code> and use these credentials in the request.</p>
<p>If the parameter being passed is an object then it can be passed as a JSON string rather than using the <code>--json</code> flag which simplifies the command line.</p>
<p>The rc interface supports some special parameters which apply to <strong>all</strong> commands. These start with <code>_</code> to show they are different.</p>
<h3id="running-asynchronous-jobs-with-_async-true">Running asynchronous jobs with _async = true</h3>
<p>Each rc call is classified as a job and it is assigned its own id. By default jobs are executed immediately as they are created or synchronously.</p>
<p>If <code>_async</code> has a true value when supplied to an rc call then it will return immediately with a job id and the task will be run in the background. The <code>job/status</code> call can be used to get information of the background job. The job can be queried for up to 1 minute after it has finished.</p>
<p>It is recommended that potentially long running jobs, e.g. <code>sync/sync</code>, <code>sync/copy</code>, <code>sync/move</code>, <code>operations/purge</code> are run with the <code>_async</code> flag to avoid any potential problems with the HTTP request and response timing out.</p>
<p>Query the status to see if the job has finished. For more information on the meaning of these return parameters see the <code>job/status</code> call.</p>
<h3id="setting-config-flags-with-_config">Setting config flags with _config</h3>
<p>If you wish to set config (the equivalent of the global flags) for the duration of an rc call only then pass in the <code>_config</code> parameter.</p>
<p>This should be in the same format as the <code>config</code> key returned by <ahref="#options-get">options/get</a>.</p>
<p>For example, if you wished to run a sync with the <code>--checksum</code> parameter, you would pass this parameter in your JSON blob.</p>
<p>Any config parameters you don't set will inherit the global defaults which were set with command line flags or environment variables.</p>
<p>Note that it is possible to set some values as strings or integers - see <ahref="/#data-types">data types</a> for more info. Here is an example setting the equivalent of <code>--buffer-size</code> in string or integer format.</p>
<p>If you wish to check the <code>_config</code> assignment has worked properly then calling <code>options/local</code> will show what the value got set to.</p>
<h3id="setting-filter-flags-with-_filter">Setting filter flags with _filter</h3>
<p>If you wish to set filters for the duration of an rc call only then pass in the <code>_filter</code> parameter.</p>
<p>This should be in the same format as the <code>filter</code> key returned by <ahref="#options-get">options/get</a>.</p>
<p>For example, if you wished to run a sync with these flags</p>
<p>Any filter parameters you don't set will inherit the global defaults which were set with command line flags or environment variables.</p>
<p>Note that it is possible to set some values as strings or integers - see <ahref="/#data-types">data types</a> for more info. Here is an example setting the equivalent of <code>--buffer-size</code> in string or integer format.</p>
<p>If you wish to check the <code>_filter</code> assignment has worked properly then calling <code>options/local</code> will show what the value got set to.</p>
<h3id="assigning-operations-to-groups-with-_group-value">Assigning operations to groups with _group = value</h3>
<p>Each rc call has its own stats group for tracking its metrics. By default grouping is done by the composite group name from prefix <code>job/</code> and id of the job like so <code>job/1</code>.</p>
<p>If <code>_group</code> has a value then stats for that request will be grouped under that value. This allows caller to group stats under their own name.</p>
<p>Stats for specific group can be accessed by passing <code>group</code> to <code>core/stats</code>:</p>
<p>When the API returns types, these will mostly be straight forward integer, string or boolean types.</p>
<p>However some of the types returned by the <ahref="#options-get">options/get</a> call and taken by the <ahref="#options-set">options/set</a> calls as well as the <code>vfsOpt</code>, <code>mountOpt</code> and the <code>_config</code> parameters.</p>
<ul>
<li><code>Duration</code> - these are returned as an integer duration in nanoseconds. They may be set as an integer, or they may be set with time string, eg "5s". See the <ahref="https://rclone.org/docs/#options">options section</a> for more info.</li>
<li><code>Size</code> - these are returned as an integer number of bytes. They may be set as an integer or they may be set with a size suffix string, eg "10M". See the <ahref="https://rclone.org/docs/#options">options section</a> for more info.</li>
<li>Enumerated type (such as <code>CutoffMode</code>, <code>DumpFlags</code>, <code>LogLevel</code>, <code>VfsCacheMode</code> - these will be returned as an integer and may be set as an integer but more conveniently they can be set as a string, eg "HARD" for <code>CutoffMode</code> or <code>DEBUG</code> for <code>LogLevel</code>.</li>
<li><code>BandwidthSpec</code> - this will be set and returned as a string, eg "1M".</li>
</ul>
<h2id="specifying-remotes-to-work-on">Specifying remotes to work on</h2>
<p>Remotes are specified with the <code>fs=</code>, <code>srcFs=</code>, <code>dstFs=</code> parameters depending on the command being used.</p>
<p>The parameters can be a string as per the rest of rclone, eg <code>s3:bucket/path</code> or <code>:sftp:/my/dir</code>. They can also be specified as JSON blobs.</p>
<p>If specifyng a JSON blob it should be a object mapping strings to strings. These values will be used to configure the remote. There are 3 special values which may be set:</p>
<ul>
<li><code>type</code> - set to <code>type</code> to specify a remote called <code>:type:</code></li>
<li><code>_name</code> - set to <code>name</code> to specify a remote called <code>name:</code></li>
<li><code>_root</code> - sets the root of the remote - may be empty</li>
</ul>
<p>One of <code>_name</code> or <code>type</code> should normally be set. If the <code>local</code> backend is desired then <code>type</code> should be set to <code>local</code>. If <code>_root</code> isn't specified then it defaults to the root of the remote.</p>
<p>For example this JSON is equivalent to <code>remote:/tmp</code></p>
<pre><code>{
"_name": "remote",
"_path": "/tmp"
}</code></pre>
<p>And this is equivalent to <code>:sftp,host='example.com':/tmp</code></p>
<pre><code>{
"type": "sftp",
"host": "example.com",
"_path": "/tmp"
}</code></pre>
<p>And this is equivalent to <code>/tmp/dir</code></p>
<p>Purge a remote from the cache backend. Supports either a directory or a file. Params: - remote = path to remote (required) - withData = true/false to delete cached data (chunks) as well (optional)</p>
<p>Ensure the specified file chunks are cached on disk.</p>
<p>The chunks= parameter specifies the file chunks to check. It takes a comma separated list of array slice indices. The slice indices are similar to Python slices: start[:end]</p>
<p>start is the 0 based chunk number from the beginning of the file to fetch inclusive. end is 0 based chunk number from the beginning of the file to fetch exclusive. Both values can be negative, in which case they count from the back of the file. The value "-5:" represents the last 5 chunks of a file.</p>
<p>Some valid examples are: ":5,-5:" -> the first and last five chunks "0,-2" -> the first and the second last chunk "0:10" -> the first ten chunks</p>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="core-gc">core/gc: Runs a garbage collection.</h3>
<p>This tells the go runtime to do a garbage collection run. It isn't necessary to call this normally, but it can be useful for debugging memory problems.</p>
<h3id="core-group-list">core/group-list: Returns list of stats.</h3>
<p>This returns list of stats groups currently in memory.</p>
<p>Returns the following values:</p>
<pre><code>{
"groups": an array of group names:
[
"group1",
"group2",
...
]
}</code></pre>
<h3id="core-memstats">core/memstats: Returns the memory statistics</h3>
<p>This returns the memory statistics of the running program. What the values mean are explained in the go docs: https://golang.org/pkg/runtime/#MemStats</p>
<p>The most interesting values for most people are:</p>
<ul>
<li>HeapAlloc: This is the amount of memory rclone is actually using</li>
<li>HeapSys: This is the amount of memory rclone has obtained from the OS</li>
<li>Sys: this is the total amount of memory requested from the OS
<ul>
<li>It is virtual memory so may include unused memory</li>
</ul></li>
</ul>
<h3id="core-obscure">core/obscure: Obscures a string passed in.</h3>
<p>Pass a clear string and rclone will obscure it for the config file: - clear - string</p>
<p>Returns - obscured - string</p>
<h3id="core-pid">core/pid: Return PID of current process</h3>
<p>This returns PID of current process. Useful for stopping rclone process.</p>
<h3id="core-quit">core/quit: Terminates the app.</h3>
<p>(optional) Pass an exit code to be used for terminating the app: - exitCode - int</p>
<h3id="core-stats">core/stats: Returns stats about current transfers.</h3>
<p>This returns all available stats:</p>
<pre><code>rclone rc core/stats</code></pre>
<p>If group is not provided then summed up stats for all groups will be returned.</p>
"transferring": an array of currently active file transfers:
[
{
"bytes": total transferred bytes for this file,
"eta": estimated time in seconds until file transfer completion
"name": name of the file,
"percentage": progress of the file transfer in percent,
"speed": average speed over the whole transfer in bytes/sec,
"speedAvg": current speed in bytes/sec as an exponentially weighted moving average,
"size": size of the file in bytes
}
],
"checking": an array of names of currently active file checks
[]
}</code></pre>
<p>Values for "transferring", "checking" and "lastError" are only assigned if data is available. The value for "eta" is null if an eta cannot be determined.</p>
<h3id="debug-set-block-profile-rate">debug/set-block-profile-rate: Set runtime.SetBlockProfileRate for blocking profiling.</h3>
<p>SetBlockProfileRate controls the fraction of goroutine blocking events that are reported in the blocking profile. The profiler aims to sample an average of one blocking event per rate nanoseconds spent blocked.</p>
<p>To include every blocking event in the profile, pass rate = 1. To turn off profiling entirely, pass rate <= 0.</p>
<p>After calling this you can use this to see the blocking profile:</p>
<h3id="debug-set-mutex-profile-fraction">debug/set-mutex-profile-fraction: Set runtime.SetMutexProfileFraction for mutex profiling.</h3>
<p>SetMutexProfileFraction controls the fraction of mutex contention events that are reported in the mutex profile. On average 1/rate events are reported. The previous rate is returned.</p>
<p>To turn off profiling entirely, pass rate 0. To just read the current rate, pass rate < 0. (For n>1 the details of sampling may change.)</p>
<p>Once this is set you can look use this to profile the mutex contention:</p>
<p>The vfsOpt are as described in options/get and can be seen in the the "vfs" section when running and the mountOpt can be seen in the "mount" section.</p>
<pre><code>rclone rc options/get</code></pre>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="mount-types">mount/types: Show all possible mount types</h3>
<p>This shows all possible mount types and returns them as a list.</p>
<p>This takes no parameters and returns</p>
<ul>
<li>mountTypes: list of mount types</li>
</ul>
<p>The mount types are strings like "mount", "mount2", "cmount" and can be passed to mount/mount as the mountType parameter.</p>
<p>Eg</p>
<pre><code>rclone rc mount/types</code></pre>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="mount-unmount">mount/unmount: Unmount selected active mount</h3>
<p>rclone allows Linux, FreeBSD, macOS and Windows to mount any of Rclone's cloud storage systems as a file system with FUSE.</p>
<p>This takes the following parameters</p>
<ul>
<li>mountPoint: valid path on the local machine where the mount was created (required)</li>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="operations-about">operations/about: Return the space used on the remote</h3>
<p>This takes the following parameters</p>
<ul>
<li>fs - a remote name string e.g. "drive:"</li>
</ul>
<p>The result is as returned from rclone about --json</p>
<p>See the <ahref="https://rclone.org/commands/rclone_size/">about command</a> command for more information on the above.</p>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="operations-cleanup">operations/cleanup: Remove trashed files in the remote or path</h3>
<p>This takes the following parameters</p>
<ul>
<li>fs - a remote name string e.g. "drive:"</li>
</ul>
<p>See the <ahref="https://rclone.org/commands/rclone_cleanup/">cleanup command</a> command for more information on the above.</p>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="operations-copyfile">operations/copyfile: Copy a file from source remote to destination remote</h3>
<p>This takes the following parameters</p>
<ul>
<li>srcFs - a remote name string e.g. "drive:" for the source</li>
<li>srcRemote - a path within that remote e.g. "file.txt" for the source</li>
<li>dstFs - a remote name string e.g. "drive2:" for the destination</li>
<li>dstRemote - a path within that remote e.g. "file2.txt" for the destination</li>
</ul>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="operations-copyurl">operations/copyurl: Copy the URL to the object</h3>
<p>This takes the following parameters</p>
<ul>
<li>fs - a remote name string e.g. "drive:"</li>
<li>remote - a path within that remote e.g. "dir"</li>
<li>url - string, URL to read from</li>
<li>autoFilename - boolean, set to true to retrieve destination file name from url See the <ahref="https://rclone.org/commands/rclone_copyurl/">copyurl command</a> command for more information on the above.</li>
</ul>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="operations-delete">operations/delete: Remove files in the path</h3>
<p>This takes the following parameters</p>
<ul>
<li>fs - a remote name string e.g. "drive:"</li>
</ul>
<p>See the <ahref="https://rclone.org/commands/rclone_delete/">delete command</a> command for more information on the above.</p>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="operations-deletefile">operations/deletefile: Remove the single file pointed to</h3>
<p>This takes the following parameters</p>
<ul>
<li>fs - a remote name string e.g. "drive:"</li>
<li>remote - a path within that remote e.g. "dir"</li>
</ul>
<p>See the <ahref="https://rclone.org/commands/rclone_deletefile/">deletefile command</a> command for more information on the above.</p>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="operations-fsinfo">operations/fsinfo: Return information about the remote</h3>
<p>This takes the following parameters</p>
<ul>
<li>fs - a remote name string e.g. "drive:"</li>
</ul>
<p>This returns info about the remote passed in;</p>
<pre><code>{
// optional features and whether they are available or not
"Features": {
"About": true,
"BucketBased": false,
"CanHaveEmptyDirectories": true,
"CaseInsensitive": false,
"ChangeNotify": false,
"CleanUp": false,
"Copy": false,
"DirCacheFlush": false,
"DirMove": true,
"DuplicateFiles": false,
"GetTier": false,
"ListR": false,
"MergeDirs": false,
"Move": true,
"OpenWriterAt": true,
"PublicLink": false,
"Purge": true,
"PutStream": true,
"PutUnchecked": false,
"ReadMimeType": false,
"ServerSideAcrossConfigs": false,
"SetTier": false,
"SetWrapper": false,
"UnWrap": false,
"WrapFs": false,
"WriteMimeType": false
},
// Names of hashes available
"Hashes": [
"MD5",
"SHA-1",
"DropboxHash",
"QuickXorHash"
],
"Name": "local", // Name as created
"Precision": 1, // Precision of timestamps in ns
"Root": "/", // Path as created
"String": "Local file system at /" // how the remote will appear in logs
}</code></pre>
<p>This command does not have a command line equivalent so use this instead:</p>
<h3id="operations-list">operations/list: List the given remote and path in JSON format</h3>
<p>This takes the following parameters</p>
<ul>
<li>fs - a remote name string e.g. "drive:"</li>
<li>remote - a path within that remote e.g. "dir"</li>
<li>opt - a dictionary of options to control the listing (optional)
<ul>
<li>recurse - If set recurse directories</li>
<li>noModTime - If set return modification time</li>
<li>showEncrypted - If set show decrypted names</li>
<li>showOrigIDs - If set show the IDs for each item if known</li>
<li>showHash - If set return a dictionary of hashes</li>
</ul></li>
</ul>
<p>The result is</p>
<ul>
<li>list
<ul>
<li>This is an array of objects as described in the lsjson command</li>
</ul></li>
</ul>
<p>See the <ahref="https://rclone.org/commands/rclone_lsjson/">lsjson command</a> for more information on the above and examples.</p>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="operations-mkdir">operations/mkdir: Make a destination directory or container</h3>
<p>This takes the following parameters</p>
<ul>
<li>fs - a remote name string e.g. "drive:"</li>
<li>remote - a path within that remote e.g. "dir"</li>
</ul>
<p>See the <ahref="https://rclone.org/commands/rclone_mkdir/">mkdir command</a> command for more information on the above.</p>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="operations-movefile">operations/movefile: Move a file from source remote to destination remote</h3>
<p>This takes the following parameters</p>
<ul>
<li>srcFs - a remote name string e.g. "drive:" for the source</li>
<li>srcRemote - a path within that remote e.g. "file.txt" for the source</li>
<li>dstFs - a remote name string e.g. "drive2:" for the destination</li>
<li>dstRemote - a path within that remote e.g. "file2.txt" for the destination</li>
</ul>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="operations-publiclink">operations/publiclink: Create or retrieve a public link to the given file or folder.</h3>
<p>This takes the following parameters</p>
<ul>
<li>fs - a remote name string e.g. "drive:"</li>
<li>remote - a path within that remote e.g. "dir"</li>
<li>unlink - boolean - if set removes the link rather than adding it (optional)</li>
<li>expire - string - the expiry time of the link e.g. "1d" (optional)</li>
</ul>
<p>Returns</p>
<ul>
<li>url - URL of the resource</li>
</ul>
<p>See the <ahref="https://rclone.org/commands/rclone_link/">link command</a> command for more information on the above.</p>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="operations-purge">operations/purge: Remove a directory or container and all of its contents</h3>
<p>This takes the following parameters</p>
<ul>
<li>fs - a remote name string e.g. "drive:"</li>
<li>remote - a path within that remote e.g. "dir"</li>
</ul>
<p>See the <ahref="https://rclone.org/commands/rclone_purge/">purge command</a> command for more information on the above.</p>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="operations-rmdir">operations/rmdir: Remove an empty directory or container</h3>
<p>This takes the following parameters</p>
<ul>
<li>fs - a remote name string e.g. "drive:"</li>
<li>remote - a path within that remote e.g. "dir"</li>
</ul>
<p>See the <ahref="https://rclone.org/commands/rclone_rmdir/">rmdir command</a> command for more information on the above.</p>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="operations-rmdirs">operations/rmdirs: Remove all the empty directories in the path</h3>
<p>This takes the following parameters</p>
<ul>
<li>fs - a remote name string e.g. "drive:"</li>
<li>remote - a path within that remote e.g. "dir"</li>
<li>leaveRoot - boolean, set to true not to delete the root</li>
</ul>
<p>See the <ahref="https://rclone.org/commands/rclone_rmdirs/">rmdirs command</a> command for more information on the above.</p>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="operations-size">operations/size: Count the number of bytes and files in remote</h3>
<p>This takes the following parameters</p>
<ul>
<li>fs - a remote name string e.g. "drive:path/to/dir"</li>
</ul>
<p>Returns</p>
<ul>
<li>count - number of files</li>
<li>bytes - number of bytes in those files</li>
</ul>
<p>See the <ahref="https://rclone.org/commands/rclone_size/">size command</a> command for more information on the above.</p>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="operations-uploadfile">operations/uploadfile: Upload file using multiform/form-data</h3>
<p>This takes the following parameters</p>
<ul>
<li>fs - a remote name string e.g. "drive:"</li>
<li>remote - a path within that remote e.g. "dir"</li>
<li>each part in body represents a file to be uploaded See the <ahref="https://rclone.org/commands/rclone_uploadfile/">uploadfile command</a> command for more information on the above.</li>
</ul>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="options-blocks">options/blocks: List all the option blocks</h3>
<p>Returns - options - a list of the options block names</p>
<p>Note that these are the global options which are unaffected by use of the _config and _filter parameters. If you wish to read the parameters set in _config then use options/config and for _filter use options/filter.</p>
<p>This shows the internal names of the option within rclone which should map to the external options very easily with a few exceptions.</p>
<h3id="options-local">options/local: Get the currently active config for this call</h3>
<p>Returns an object with the keys "config" and "filter". The "config" key contains the local config and the "filter" key contains the local filters.</p>
<p>Note that these are the local options specific to this rc call. If _config was not supplied then they will be the global options. Likewise with "_filter".</p>
<p>This call is mostly useful for seeing if _config and _filter passing is working.</p>
<p>This shows the internal names of the option within rclone which should map to the external options very easily with a few exceptions.</p>
<h3id="options-set">options/set: Set an option</h3>
<p>Parameters</p>
<ul>
<li>option block name containing an object with
<ul>
<li>key: value</li>
</ul></li>
</ul>
<p>Repeated as often as required.</p>
<p>Only supply the options you wish to change. If an option is unknown it will be silently ignored. Not all options will have an effect when changed like this.</p>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="rc-error">rc/error: This returns an error</h3>
<p>This returns an error with the input as part of its error string. Useful for testing error handling.</p>
<h3id="rc-list">rc/list: List all the registered remote control commands</h3>
<p>This lists all the registered remote control commands as a JSON map in the commands response.</p>
<h3id="rc-noop">rc/noop: Echo the input to the output parameters</h3>
<p>This echoes the input parameters to the output parameters for testing purposes. It can be used to check that rclone is still alive and to check that parameter passing is working properly.</p>
<h3id="rc-noopauth">rc/noopauth: Echo the input to the output parameters requiring auth</h3>
<p>This echoes the input parameters to the output parameters for testing purposes. It can be used to check that rclone is still alive and to check that parameter passing is working properly.</p>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="sync-copy">sync/copy: copy a directory from source remote to destination remote</h3>
<p>This takes the following parameters</p>
<ul>
<li>srcFs - a remote name string e.g. "drive:src" for the source</li>
<li>dstFs - a remote name string e.g. "drive:dst" for the destination</li>
<p>See the <ahref="https://rclone.org/commands/rclone_sync/">sync command</a> command for more information on the above.</p>
<p><strong>Authentication is required for this call.</strong></p>
<h3id="vfs-forget">vfs/forget: Forget files or directories in the directory cache.</h3>
<p>This forgets the paths in the directory cache causing them to be re-read from the remote when needed.</p>
<p>If no paths are passed in then it will forget all the paths in the directory cache.</p>
<pre><code>rclone rc vfs/forget</code></pre>
<p>Otherwise pass files or dirs in as file=path or dir=path. Any parameter key starting with file will forget that file and any starting with dir will forget that dir, e.g.</p>
<p>This command takes an "fs" parameter. If this parameter is not supplied and if there is only one VFS in use then that VFS will be used. If there is more than one VFS in use then the "fs" parameter must be supplied.</p>
<h3id="vfs-list">vfs/list: List active VFSes.</h3>
<p>This lists the active VFSes.</p>
<p>It returns a list under the key "vfses" where the values are the VFS names that could be passed to the other VFS commands in the "fs" parameter.</p>
<h3id="vfs-poll-interval">vfs/poll-interval: Get the status or update the value of the poll-interval option.</h3>
<p>Without any parameter given this returns the current status of the poll-interval setting.</p>
<p>When the interval=duration parameter is set, the poll-interval value is updated and the polling function is notified. Setting interval=0 disables poll-interval.</p>
<p>The timeout=duration parameter can be used to specify a time to wait for the current poll function to apply the new value. If timeout is less or equal 0, which is the default, wait indefinitely.</p>
<p>The new poll-interval value will only be active when the timeout is not reached.</p>
<p>If poll-interval is updated or disabled temporarily, some changes might not get picked up by the polling function, depending on the used remote.</p>
<p>This command takes an "fs" parameter. If this parameter is not supplied and if there is only one VFS in use then that VFS will be used. If there is more than one VFS in use then the "fs" parameter must be supplied.</p>
<h3id="vfs-refresh">vfs/refresh: Refresh the directory cache.</h3>
<p>This reads the directories for the specified paths and freshens the directory cache.</p>
<p>If no paths are passed in then it will refresh the root directory.</p>
<pre><code>rclone rc vfs/refresh</code></pre>
<p>Otherwise pass directories in as dir=path. Any parameter key starting with dir will refresh that directory, e.g.</p>
<p>If the parameter recursive=true is given the whole directory tree will get refreshed. This refresh will use --fast-list if enabled.</p>
<p>This command takes an "fs" parameter. If this parameter is not supplied and if there is only one VFS in use then that VFS will be used. If there is more than one VFS in use then the "fs" parameter must be supplied.</p>
<h2id="api-http">Accessing the remote control via HTTP</h2>
<p>Rclone implements a simple HTTP based protocol.</p>
<p>Each endpoint takes an JSON object and returns a JSON object or an error. The JSON objects are essentially a map of string names to values.</p>
<p>All calls must made using POST.</p>
<p>The input objects can be supplied using URL parameters, POST parameters or by supplying "Content-Type: application/json" and a JSON blob in the body. There are examples of these below using <code>curl</code>.</p>
<p>The response will be a JSON blob in the body of the response. This is formatted to be reasonably human readable.</p>
<h3id="error-returns">Error returns</h3>
<p>If an error occurs then there will be an HTTP error status (e.g. 500) and the body of the response will contain a JSON encoded error object, e.g.</p>
<pre><code>{
"error": "Expecting string value for key \"remote\" (was float64)",
"input": {
"fs": "/tmp",
"remote": 3
},
"status": 400
"path": "operations/rmdir",
}</code></pre>
<p>The keys in the error response are - error - error string - input - the input parameters to the call - status - the HTTP status code - path - the path of the call</p>
<h3id="cors">CORS</h3>
<p>The sever implements basic CORS support and allows all origins for that. The response to a preflight OPTIONS request will echo the requested "Access-Control-Request-Headers" back.</p>
<h3id="using-post-with-url-parameters-only">Using POST with URL parameters only</h3>
<pre><code>curl -X POST 'http://localhost:5572/rc/noop?potato=1&sausage=2'</code></pre>
<p>Response</p>
<pre><code>{
"potato": "1",
"sausage": "2"
}</code></pre>
<p>Here is what an error response looks like:</p>
<pre><code>curl -X POST 'http://localhost:5572/rc/error?potato=1&sausage=2'</code></pre>
<pre><code>{
"error": "arbitrary error on input map[potato:1 sausage:2]",
"input": {
"potato": "1",
"sausage": "2"
}
}</code></pre>
<p>Note that curl doesn't return errors to the shell unless you use the <code>-f</code> option</p>
<pre><code>$ curl -f -X POST 'http://localhost:5572/rc/error?potato=1&sausage=2'
curl: (22) The requested URL returned error: 400 Bad Request
$ echo $?
22</code></pre>
<h3id="using-post-with-a-form">Using POST with a form</h3>
<p>See the <ahref="https://golang.org/pkg/net/http/pprof/">net/http/pprof docs</a> for more info on how to use the profiling and for a general overview see <ahref="https://blog.golang.org/profiling-go-programs">the Go team's blog post on profiling go programs</a>.</p>
<p>The profiling hook is <ahref="https://stackoverflow.com/q/26545159/164234">zero overhead unless it is used</a>.</p>
<h1id="overview-of-cloud-storage-systems">Overview of cloud storage systems</h1>
<p>Each cloud storage system is slightly different. Rclone attempts to provide a unified interface to them, but some underlying differences show through.</p>
<h2id="features-1">Features</h2>
<p>Here is an overview of the major features of each cloud storage system.</p>
<p>¹ Dropbox supports <ahref="https://www.dropbox.com/developers/reference/content-hash">its own custom hash</a>. This is an SHA256 sum of all the 4MB block SHA256s.</p>
<p>² SFTP supports checksums if the same login has shell access and <code>md5sum</code> or <code>sha1sum</code> as well as <code>echo</code> are in the remote's PATH.</p>
<p>³ WebDAV supports hashes when used with Owncloud and Nextcloud only.</p>
<p>⁴ WebDAV supports modtimes when used with Owncloud and Nextcloud only.</p>
<p>⁵ Microsoft OneDrive Personal supports SHA1 hashes, whereas OneDrive for business and SharePoint server support Microsoft's own <ahref="https://docs.microsoft.com/en-us/onedrive/developer/code-snippets/quickxorhash">QuickXorHash</a>.</p>
<p>⁶ Mail.ru uses its own modified SHA1 hash</p>
<p>⁷ pCloud only supports SHA1 (not MD5) in its EU region</p>
<p>⁸ Opendrive does not support creation of duplicate files using their web client interface or other stock clients, but the underlying storage platform has been determined to allow duplicate files, and it is possible to create them with <code>rclone</code>. It may be that this is a mistake or an unsupported feature.</p>
<h3id="hash">Hash</h3>
<p>The cloud storage system supports various hash types of the objects. The hashes are used when transferring data as an integrity check and can be specifically used with the <code>--checksum</code> flag in syncs and in the <code>check</code> command.</p>
<p>To use the verify checksums when transferring between cloud storage systems they must support a common hash type.</p>
<h3id="modtime">ModTime</h3>
<p>The cloud storage system supports setting modification times on objects. If it does then this enables a using the modification times as part of the sync. If not then only the size will be checked by default, though the MD5SUM can be checked with the <code>--checksum</code> flag.</p>
<p>All cloud storage systems support some kind of date on the object and these will be set when transferring from the cloud storage system.</p>
<h3id="case-insensitive">Case Insensitive</h3>
<p>If a cloud storage systems is case sensitive then it is possible to have two files which differ only in case, e.g. <code>file.txt</code> and <code>FILE.txt</code>. If a cloud storage system is case insensitive then that isn't possible.</p>
<p>This can cause problems when syncing between a case insensitive system and a case sensitive system. The symptom of this is that no matter how many times you run the sync it never completes fully.</p>
<p>The local filesystem and SFTP may or may not be case sensitive depending on OS.</p>
<ul>
<li>Windows - usually case insensitive, though case is preserved</li>
<li>OSX - usually case insensitive, though it is possible to format case sensitive</li>
<li>Linux - usually case sensitive, but there are case insensitive file systems (e.g. FAT formatted USB keys)</li>
</ul>
<p>Most of the time this doesn't cause any problems as people tend to avoid files whose name differs only by case even on case sensitive systems.</p>
<h3id="duplicate-files">Duplicate files</h3>
<p>If a cloud storage system allows duplicate files then it can have two objects with the same name.</p>
<p>This confuses rclone greatly when syncing - use the <code>rclone dedupe</code> command to rename or remove duplicates.</p>
<p>Some cloud storage systems might have restrictions on the characters that are usable in file or directory names. When <code>rclone</code> detects such a name during a file upload, it will transparently replace the restricted characters with similar looking Unicode characters.</p>
<p>This process is designed to avoid ambiguous file names as much as possible and allow to move files between many cloud storage systems transparently.</p>
<p>The name shown by <code>rclone</code> to the user or during log output will only contain a minimal set of <ahref="#restricted-characters">replaced characters</a> to ensure correct formatting and not necessarily the actual name used on the cloud storage.</p>
<p>This transformation is reversed when downloading a file or parsing <code>rclone</code> arguments. For example, when uploading a file named <code>my file?.txt</code> to Onedrive will be displayed as <code>my file?.txt</code> on the console, but stored as <code>my file？.txt</code> (the <code>?</code> gets replaced by the similar looking <code>？</code> character) to Onedrive. The reverse transformation allows to read a file<code>unusual/name.txt</code> from Google Drive, by passing the name <code>unusual／name.txt</code> (the <code>/</code> needs to be replaced by the similar looking <code>／</code> character) on the command line.</p>
<p>The table below shows the characters that are replaced by default.</p>
<p>When a replacement character is found in a filename, this character will be escaped with the <code>‛</code> character to avoid ambiguous file names. (e.g. a file named <code>␀.txt</code> would shown as <code>‛␀.txt</code>)</p>
<p>Each cloud storage backend can use a different set of characters, which will be specified in the documentation for each backend.</p>
<table>
<thead>
<trclass="header">
<th>Character</th>
<thstyle="text-align: center;">Value</th>
<thstyle="text-align: center;">Replacement</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>NUL</td>
<tdstyle="text-align: center;">0x00</td>
<tdstyle="text-align: center;">␀</td>
</tr>
<trclass="even">
<td>SOH</td>
<tdstyle="text-align: center;">0x01</td>
<tdstyle="text-align: center;">␁</td>
</tr>
<trclass="odd">
<td>STX</td>
<tdstyle="text-align: center;">0x02</td>
<tdstyle="text-align: center;">␂</td>
</tr>
<trclass="even">
<td>ETX</td>
<tdstyle="text-align: center;">0x03</td>
<tdstyle="text-align: center;">␃</td>
</tr>
<trclass="odd">
<td>EOT</td>
<tdstyle="text-align: center;">0x04</td>
<tdstyle="text-align: center;">␄</td>
</tr>
<trclass="even">
<td>ENQ</td>
<tdstyle="text-align: center;">0x05</td>
<tdstyle="text-align: center;">␅</td>
</tr>
<trclass="odd">
<td>ACK</td>
<tdstyle="text-align: center;">0x06</td>
<tdstyle="text-align: center;">␆</td>
</tr>
<trclass="even">
<td>BEL</td>
<tdstyle="text-align: center;">0x07</td>
<tdstyle="text-align: center;">␇</td>
</tr>
<trclass="odd">
<td>BS</td>
<tdstyle="text-align: center;">0x08</td>
<tdstyle="text-align: center;">␈</td>
</tr>
<trclass="even">
<td>HT</td>
<tdstyle="text-align: center;">0x09</td>
<tdstyle="text-align: center;">␉</td>
</tr>
<trclass="odd">
<td>LF</td>
<tdstyle="text-align: center;">0x0A</td>
<tdstyle="text-align: center;">␊</td>
</tr>
<trclass="even">
<td>VT</td>
<tdstyle="text-align: center;">0x0B</td>
<tdstyle="text-align: center;">␋</td>
</tr>
<trclass="odd">
<td>FF</td>
<tdstyle="text-align: center;">0x0C</td>
<tdstyle="text-align: center;">␌</td>
</tr>
<trclass="even">
<td>CR</td>
<tdstyle="text-align: center;">0x0D</td>
<tdstyle="text-align: center;">␍</td>
</tr>
<trclass="odd">
<td>SO</td>
<tdstyle="text-align: center;">0x0E</td>
<tdstyle="text-align: center;">␎</td>
</tr>
<trclass="even">
<td>SI</td>
<tdstyle="text-align: center;">0x0F</td>
<tdstyle="text-align: center;">␏</td>
</tr>
<trclass="odd">
<td>DLE</td>
<tdstyle="text-align: center;">0x10</td>
<tdstyle="text-align: center;">␐</td>
</tr>
<trclass="even">
<td>DC1</td>
<tdstyle="text-align: center;">0x11</td>
<tdstyle="text-align: center;">␑</td>
</tr>
<trclass="odd">
<td>DC2</td>
<tdstyle="text-align: center;">0x12</td>
<tdstyle="text-align: center;">␒</td>
</tr>
<trclass="even">
<td>DC3</td>
<tdstyle="text-align: center;">0x13</td>
<tdstyle="text-align: center;">␓</td>
</tr>
<trclass="odd">
<td>DC4</td>
<tdstyle="text-align: center;">0x14</td>
<tdstyle="text-align: center;">␔</td>
</tr>
<trclass="even">
<td>NAK</td>
<tdstyle="text-align: center;">0x15</td>
<tdstyle="text-align: center;">␕</td>
</tr>
<trclass="odd">
<td>SYN</td>
<tdstyle="text-align: center;">0x16</td>
<tdstyle="text-align: center;">␖</td>
</tr>
<trclass="even">
<td>ETB</td>
<tdstyle="text-align: center;">0x17</td>
<tdstyle="text-align: center;">␗</td>
</tr>
<trclass="odd">
<td>CAN</td>
<tdstyle="text-align: center;">0x18</td>
<tdstyle="text-align: center;">␘</td>
</tr>
<trclass="even">
<td>EM</td>
<tdstyle="text-align: center;">0x19</td>
<tdstyle="text-align: center;">␙</td>
</tr>
<trclass="odd">
<td>SUB</td>
<tdstyle="text-align: center;">0x1A</td>
<tdstyle="text-align: center;">␚</td>
</tr>
<trclass="even">
<td>ESC</td>
<tdstyle="text-align: center;">0x1B</td>
<tdstyle="text-align: center;">␛</td>
</tr>
<trclass="odd">
<td>FS</td>
<tdstyle="text-align: center;">0x1C</td>
<tdstyle="text-align: center;">␜</td>
</tr>
<trclass="even">
<td>GS</td>
<tdstyle="text-align: center;">0x1D</td>
<tdstyle="text-align: center;">␝</td>
</tr>
<trclass="odd">
<td>RS</td>
<tdstyle="text-align: center;">0x1E</td>
<tdstyle="text-align: center;">␞</td>
</tr>
<trclass="even">
<td>US</td>
<tdstyle="text-align: center;">0x1F</td>
<tdstyle="text-align: center;">␟</td>
</tr>
<trclass="odd">
<td>/</td>
<tdstyle="text-align: center;">0x2F</td>
<tdstyle="text-align: center;">／</td>
</tr>
<trclass="even">
<td>DEL</td>
<tdstyle="text-align: center;">0x7F</td>
<tdstyle="text-align: center;">␡</td>
</tr>
</tbody>
</table>
<p>The default encoding will also encode these file names as they are problematic with many cloud storage systems.</p>
<table>
<thead>
<trclass="header">
<th>File name</th>
<thstyle="text-align: center;">Replacement</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>.</td>
<tdstyle="text-align: center;">．</td>
</tr>
<trclass="even">
<td>..</td>
<tdstyle="text-align: center;">．．</td>
</tr>
</tbody>
</table>
<h4id="invalid-utf8">Invalid UTF-8 bytes</h4>
<p>Some backends only support a sequence of well formed UTF-8 bytes as file or directory names.</p>
<p>In this case all invalid UTF-8 bytes will be replaced with a quoted representation of the byte value to allow uploading a file to such a backend. For example, the invalid byte <code>0xFE</code> will be encoded as <code>‛FE</code>.</p>
<p>A common source of invalid UTF-8 bytes are local filesystems, that store names in a different encoding than UTF-8 or UTF-16, like latin1. See the <ahref="https://rclone.org/local/#filenames">local filenames</a> section for details.</p>
<h4id="encoding">Encoding option</h4>
<p>Most backends have an encoding options, specified as a flag <code>--backend-encoding</code> where <code>backend</code> is the name of the backend, or as a config parameter <code>encoding</code> (you'll need to select the Advanced config in <code>rclone config</code> to see it).</p>
<p>This will have default value which encodes and decodes characters in such a way as to preserve the maximum number of characters (see above).</p>
<p>However this can be incorrect in some scenarios, for example if you have a Windows file system with characters such as <code>＊</code> and <code>？</code> that you want to remain as those characters on the remote rather than being translated to <code>*</code> and <code>?</code>.</p>
<p>The <code>--backend-encoding</code> flags allow you to change that. You can disable the encoding completely with <code>--backend-encoding None</code> or set <code>encoding = None</code> in the config file.</p>
<p>Encoding takes a comma separated list of encodings. You can see the list of all available characters by passing an invalid value to this flag, e.g. <code>--local-encoding "help"</code> and <code>rclone help flags encoding</code> will show you the defaults for the backends.</p>
<table>
<thead>
<trclass="header">
<th>Encoding</th>
<th>Characters</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>Asterisk</td>
<td><code>*</code></td>
</tr>
<trclass="even">
<td>BackQuote</td>
<td><code>`</code></td>
</tr>
<trclass="odd">
<td>BackSlash</td>
<td><code>\</code></td>
</tr>
<trclass="even">
<td>Colon</td>
<td><code>:</code></td>
</tr>
<trclass="odd">
<td>CrLf</td>
<td>CR 0x0D, LF 0x0A</td>
</tr>
<trclass="even">
<td>Ctl</td>
<td>All control characters 0x00-0x1F</td>
</tr>
<trclass="odd">
<td>Del</td>
<td>DEL 0x7F</td>
</tr>
<trclass="even">
<td>Dollar</td>
<td><code>$</code></td>
</tr>
<trclass="odd">
<td>Dot</td>
<td><code>.</code></td>
</tr>
<trclass="even">
<td>DoubleQuote</td>
<td><code>"</code></td>
</tr>
<trclass="odd">
<td>Hash</td>
<td><code>#</code></td>
</tr>
<trclass="even">
<td>InvalidUtf8</td>
<td>An invalid UTF-8 character (e.g. latin1)</td>
</tr>
<trclass="odd">
<td>LeftCrLfHtVt</td>
<td>CR 0x0D, LF 0x0A,HT 0x09, VT 0x0B on the left of a string</td>
</tr>
<trclass="even">
<td>LeftPeriod</td>
<td><code>.</code> on the left of a string</td>
</tr>
<trclass="odd">
<td>LeftSpace</td>
<td>SPACE on the left of a string</td>
</tr>
<trclass="even">
<td>LeftTilde</td>
<td><code>~</code> on the left of a string</td>
</tr>
<trclass="odd">
<td>LtGt</td>
<td><code><</code>, <code>></code></td>
</tr>
<trclass="even">
<td>None</td>
<td>No characters are encoded</td>
</tr>
<trclass="odd">
<td>Percent</td>
<td><code>%</code></td>
</tr>
<trclass="even">
<td>Pipe</td>
<td>|</td>
</tr>
<trclass="odd">
<td>Question</td>
<td><code>?</code></td>
</tr>
<trclass="even">
<td>RightCrLfHtVt</td>
<td>CR 0x0D, LF 0x0A, HT 0x09, VT 0x0B on the right of a string</td>
</tr>
<trclass="odd">
<td>RightPeriod</td>
<td><code>.</code> on the right of a string</td>
</tr>
<trclass="even">
<td>RightSpace</td>
<td>SPACE on the right of a string</td>
</tr>
<trclass="odd">
<td>SingleQuote</td>
<td><code>'</code></td>
</tr>
<trclass="even">
<td>Slash</td>
<td><code>/</code></td>
</tr>
</tbody>
</table>
<p>To take a specific example, the FTP backend's default encoding is</p>
<p>However, let's say the FTP server is running on Windows and can't have any of the invalid Windows characters in file names. You are backing up Linux servers to this FTP server which do have those characters in file names. So you would add the Windows set which are</p>
<p>This can be specified using the <code>--ftp-encoding</code> flag or using an <code>encoding</code> parameter in the config file.</p>
<p>Or let's say you have a Windows server but you want to preserve <code>＊</code> and <code>？</code>, you would then have this as the encoding (the Windows encoding minus <code>Asterisk</code> and <code>Question</code>).</p>
<p>This can be specified using the <code>--local-encoding</code> flag or using an <code>encoding</code> parameter in the config file.</p>
<h3id="mime-type">MIME Type</h3>
<p>MIME types (also known as media types) classify types of documents using a simple text classification, e.g. <code>text/html</code> or <code>application/pdf</code>.</p>
<p>Some cloud storage systems support reading (<code>R</code>) the MIME type of objects and some support writing (<code>W</code>) the MIME type of objects.</p>
<p>The MIME type can be important if you are serving files directly to HTTP from the storage system.</p>
<p>If you are copying from a remote which supports reading (<code>R</code>) to a remote which supports writing (<code>W</code>) then rclone will preserve the MIME types. Otherwise they will be guessed from the extension, or the remote itself may assign the MIME type.</p>
<h2id="optional-features">Optional Features</h2>
<p>All rclone remotes support a base command set. Other features depend upon backend specific capabilities.</p>
<p>This deletes a directory quicker than just deleting all the files in the directory.</p>
<p>† Note Swift, Hubic, and Tardigrade implement this in order to delete directory markers but they don't actually have a quicker way of deleting files other than deleting them individually.</p>
<p>‡ StreamUpload is not supported with Nextcloud</p>
<h3id="copy">Copy</h3>
<p>Used when copying an object to and from the same remote. This known as a server-side copy so you can copy a file without downloading it and uploading it again. It is used if you use <code>rclone copy</code> or <code>rclone move</code> if the remote doesn't support <code>Move</code> directly.</p>
<p>If the server doesn't support <code>Copy</code> directly then for copy operations the file is downloaded then re-uploaded.</p>
<h3id="move">Move</h3>
<p>Used when moving/renaming an object on the same remote. This is known as a server-side move of a file. This is used in <code>rclone move</code> if the server doesn't support <code>DirMove</code>.</p>
<p>If the server isn't capable of <code>Move</code> then rclone simulates it with <code>Copy</code> then delete. If the server doesn't support <code>Copy</code> then rclone will download the file and re-upload it.</p>
<h3id="dirmove">DirMove</h3>
<p>This is used to implement <code>rclone move</code> to move a directory if possible. If it isn't then it will use <code>Move</code> on each file (which falls back to <code>Copy</code> then download and upload - see <code>Move</code> section).</p>
<h3id="cleanup">CleanUp</h3>
<p>This is used for emptying the trash for a remote by <code>rclone cleanup</code>.</p>
<p>If the server can't do <code>CleanUp</code> then <code>rclone cleanup</code> will return an error.</p>
<p>‡‡ Note that while Box implements this it has to delete every file individually so it will be slower than emptying the trash via the WebUI</p>
<h3id="listr">ListR</h3>
<p>The remote supports a recursive list to list all the contents beneath a directory quickly. This enables the <code>--fast-list</code> flag to work. See the <ahref="https://rclone.org/docs/#fast-list">rclone docs</a> for more details.</p>
<h3id="streamupload">StreamUpload</h3>
<p>Some remotes allow files to be uploaded without knowing the file size in advance. This allows certain operations to work without spooling the file to local disk first, e.g. <code>rclone rcat</code>.</p>
<h3id="linksharing">LinkSharing</h3>
<p>Sets the necessary permissions on a file or folder and prints a link that allows others to access them, even if they don't have an account on the particular cloud provider.</p>
<h3id="about-1">About</h3>
<p>Rclone <code>about</code> prints quota information for a remote. Typical output includes bytes used, free, quota and in trash.</p>
<p>If a remote lacks about capability <code>rclone about remote:</code>returns an error.</p>
<p>Backends without about capability cannot determine free space for an rclone mount, or use policy <code>mfs</code> (most free space) as a member of an rclone union remote.</p>
<p>See <ahref="https://rclone.org/commands/rclone_about/">rclone about command</a></p>
<h3id="emptydir">EmptyDir</h3>
<p>The remote supports empty directories. See <ahref="https://rclone.org/bugs/#limitations">Limitations</a> for details. Most Object/Bucket based remotes do not support this.</p>
<h1id="global-flags">Global Flags</h1>
<p>This describes the global flags available to every rclone command split into two groups, non backend and backend flags.</p>
<h2id="non-backend-flags">Non Backend Flags</h2>
<p>These flags are available for every command.</p>
<pre><code> --ask-password Allow prompt for password for encrypted configuration. (default true)
--auto-confirm If enabled, do not request console confirmation.
--backup-dir string Make backups into hierarchy based in DIR.
--bind string Local address to bind to for outgoing connections, IPv4, IPv6 or name.
--buffer-size SizeSuffix In memory buffer size when reading files for each --transfer. (default 16M)
--bwlimit BwTimetable Bandwidth limit in kBytes/s, or use suffix b|k|M|G or a full timetable.
--bwlimit-file BwTimetable Bandwidth limit per file in kBytes/s, or use suffix b|k|M|G or a full timetable.
--ca-cert string CA certificate used to verify servers
--cache-dir string Directory rclone will use for caching. (default "$HOME/.cache/rclone")
--check-first Do all the checks before starting transfers.
--checkers int Number of checkers to run in parallel. (default 8)
-c, --checksum Skip based on checksum (if available) & size, not mod-time & size
--client-cert string Client SSL certificate (PEM) for mutual TLS auth
--log-systemd Activate systemd integration for the logger.
--low-level-retries int Number of low level retries to do. (default 10)
--max-age Duration Only transfer files younger than this in s or suffix ms|s|m|h|d|w|M|y (default off)
--max-backlog int Maximum number of objects in sync or check backlog. (default 10000)
--max-delete int When synchronizing, limit the number of deletes (default -1)
--max-depth int If set limits the recursion depth to this. (default -1)
--max-duration duration Maximum duration rclone will transfer data for.
--max-size SizeSuffix Only transfer files smaller than this in k or suffix b|k|M|G (default off)
--max-stats-groups int Maximum number of stats groups to keep in memory. On max oldest is discarded. (default 1000)
--max-transfer SizeSuffix Maximum size of data to transfer. (default off)
--memprofile string Write memory profile to file
--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)
--min-size SizeSuffix Only transfer files bigger than this in k or suffix b|k|M|G (default off)
--modify-window duration Max time diff to be considered the same (default 1ns)
--multi-thread-cutoff SizeSuffix Use multi-thread downloads for files above this size. (default 250M)
--multi-thread-streams int Max number of streams to use for multi-thread downloads. (default 4)
--no-check-certificate Do not verify the server SSL certificate. Insecure.
--no-check-dest Don't check the destination, copy regardless.
--no-console Hide console window. Supported on Windows only.
--no-gzip-encoding Don't set Accept-Encoding: gzip.
--no-traverse Don't traverse destination file system on copy.
--no-unicode-normalization Don't normalize unicode characters in filenames.
--no-update-modtime Don't update destination mod-time if files identical.
--order-by string Instructions on how to order the transfers, e.g. 'size,descending'
--password-command SpaceSepList Command for supplying password for encrypted configuration.
-P, --progress Show progress during transfer.
--progress-terminal-title Show progress on the terminal title. Requires -P/--progress.
-q, --quiet Print as little stuff as possible
--rc Enable the remote control server.
--rc-addr string IPaddress:Port or :Port to bind server to. (default "localhost:5572")
--rc-allow-origin string Set the allowed origin for CORS.
--rc-baseurl string Prefix for URLs - leave blank for root.
--rc-cert string SSL PEM key (concatenation of certificate and CA certificate)
--rc-client-ca string Client certificate authority to verify clients with
--rc-enable-metrics Enable prometheus metrics on /metrics
--rc-files string Path to local files to serve on the HTTP server.
--rc-htpasswd string htpasswd file - if not provided no authentication is done
--rc-job-expire-duration duration expire finished async jobs older than this value (default 1m0s)
--rc-job-expire-interval duration interval to check for expired async jobs (default 10s)
--rc-key string SSL PEM Private key
--rc-max-header-bytes int Maximum size of request header (default 4096)
--rc-no-auth Don't require auth for certain methods.
--rc-pass string Password for authentication.
--rc-realm string realm for authentication (default "rclone")
--rc-serve Enable the serving of remote objects.
--rc-server-read-timeout duration Timeout for server reading data (default 1h0m0s)
--rc-server-write-timeout duration Timeout for server writing data (default 1h0m0s)
--rc-template string User Specified Template.
--rc-user string User name for authentication.
--rc-web-fetch-url string URL to fetch the releases for webgui. (default "https://api.github.com/repos/rclone/rclone-webui-react/releases/latest")
--rc-web-gui Launch WebGUI on localhost
--rc-web-gui-force-update Force update to latest version of web gui
--rc-web-gui-no-open-browser Don't open the browser automatically
--rc-web-gui-update Check and update to latest version of web gui
--refresh-times Refresh the modtime of remote files.
--retries int Retry operations this many times if they fail (default 3)
--retries-sleep duration Interval between retrying operations if they fail, e.g 500ms, 60s, 5m. (0 to disable)
--size-only Skip based on size only, not mod-time or checksum
--stats duration Interval between printing stats, e.g 500ms, 60s, 5m. (0 to disable) (default 1m0s)
--stats-file-name-length int Max file name length in stats. 0 for no limit (default 45)
--stats-log-level string Log level to show --stats output DEBUG|INFO|NOTICE|ERROR (default "INFO")
--stats-one-line Make the stats fit on one line.
--stats-one-line-date Enables --stats-one-line and add current date/time prefix.
--stats-one-line-date-format string Enables --stats-one-line-date and uses custom formatted date. Enclose date string in double quotes ("). See https://golang.org/pkg/time/#Time.Format
--stats-unit string Show data rate in stats as either 'bits' or 'bytes'/s (default "bytes")
--streaming-upload-cutoff SizeSuffix Cutoff for switching to chunked upload if file size is unknown. Upload starts after reaching cutoff or when file ends. (default 100k)
--suffix string Suffix to add to changed files.
--suffix-keep-extension Preserve the extension when using --suffix.
--syslog Use Syslog for logging
--syslog-facility string Facility for syslog, e.g. KERN,USER,... (default "DAEMON")
--timeout duration IO idle timeout (default 5m0s)
--tpslimit float Limit HTTP transactions per second to this.
--tpslimit-burst int Max burst of transactions for --tpslimit. (default 1)
--track-renames When synchronizing, track file renames and do a server-side move if possible
--track-renames-strategy string Strategies to use when synchronizing using track-renames hash|modtime|leaf (default "hash")
--transfers int Number of file transfers to run in parallel. (default 4)
-u, --update Skip files that are newer on the destination.
--use-cookies Enable session cookiejar.
--use-json-log Use json log format.
--use-mmap Use mmap allocator (see docs).
--use-server-modtime Use server modified time instead of object metadata
--crypt-password string Password or pass phrase for encryption. (obscured)
--crypt-password2 string Password or pass phrase for salt. Optional but recommended. (obscured)
--crypt-remote string Remote to encrypt/decrypt.
--crypt-server-side-across-configs Allow server-side operations (e.g. copy) to work across different crypt configs.
--crypt-show-mapping For all files listed show how the names encrypt.
--drive-acknowledge-abuse Set to allow files which return cannotDownloadAbusiveFile to be downloaded.
--drive-allow-import-name-change Allow the filetype to change when uploading Google docs (e.g. file.doc to file.docx). This will confuse sync and reupload every time.
--drive-auth-owner-only Only consider files owned by the authenticated user.
--drive-auth-url string Auth server URL.
--drive-chunk-size SizeSuffix Upload chunk size. Must a power of 2 >= 256k. (default 8M)
--drive-client-id string Google Application Client Id
--drive-client-secret string OAuth Client Secret
--drive-disable-http2 Disable drive using http2 (default true)
--drive-encoding MultiEncoder This sets the encoding for the backend. (default InvalidUtf8)
--drive-export-formats string Comma separated list of preferred formats for downloading Google docs. (default "docx,xlsx,pptx,svg")
--drive-formats string Deprecated: see export_formats
--drive-impersonate string Impersonate this user when using a service account.
--drive-import-formats string Comma separated list of preferred formats for uploading Google docs.
--drive-keep-revision-forever Keep new head revision of each file forever.
--drive-list-chunk int Size of listing chunk 100-1000. 0 to disable. (default 1000)
--drive-pacer-burst int Number of API calls to allow without sleeping. (default 100)
--drive-pacer-min-sleep Duration Minimum time to sleep between API calls. (default 100ms)
--drive-root-folder-id string ID of the root folder
--drive-scope string Scope that rclone should use when requesting access from drive.
--drive-server-side-across-configs Allow server-side operations (e.g. copy) to work across different drive configs.
--drive-service-account-credentials string Service Account Credentials JSON blob
--drive-service-account-file string Service Account Credentials JSON file path
--drive-shared-with-me Only show files that are shared with me.
--drive-size-as-quota Show sizes as storage quota usage, not actual size.
--drive-skip-checksum-gphotos Skip MD5 checksum on Google photos and videos only.
--drive-skip-gdocs Skip google documents in all listings.
--drive-skip-shortcuts If set skip shortcut files
--drive-starred-only Only show files that are starred.
--drive-stop-on-download-limit Make download limit errors be fatal
--drive-stop-on-upload-limit Make upload limit errors be fatal
--dropbox-encoding MultiEncoder This sets the encoding for the backend. (default Slash,BackSlash,Del,RightSpace,InvalidUtf8,Dot)
--dropbox-impersonate string Impersonate this user when using a business account.
--dropbox-shared-files Instructs rclone to work on individual shared files.
--dropbox-shared-folders Instructs rclone to work on shared folders.
--dropbox-token string OAuth Access Token as a JSON blob.
--dropbox-token-url string Token server url.
--fichier-api-key string Your API Key, get it from https://1fichier.com/console/params.pl
--fichier-encoding MultiEncoder This sets the encoding for the backend. (default Slash,LtGt,DoubleQuote,SingleQuote,BackQuote,Dollar,BackSlash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot)
--fichier-shared-folder string If you want to download a shared folder, add this parameter
--filefabric-encoding MultiEncoder This sets the encoding for the backend. (default Slash,Del,Ctl,InvalidUtf8,Dot)
--gphotos-include-archived Also view and download archived media.
--gphotos-read-only Set to make the Google Photos backend read only.
--gphotos-read-size Set to read the size of media items.
--gphotos-start-year int Year limits the photos to be downloaded to those which are uploaded after the given year (default 2000)
--gphotos-token string OAuth Access Token as a JSON blob.
--gphotos-token-url string Token server url.
--hdfs-data-transfer-protection string Kerberos data transfer protection: authentication|integrity|privacy
--hdfs-encoding MultiEncoder This sets the encoding for the backend. (default Slash,Colon,Del,Ctl,InvalidUtf8,Dot)
--hdfs-namenode string hadoop name node and port
--hdfs-service-principal-name string Kerberos service principal name for the namenode
--hdfs-username string hadoop user name
--http-headers CommaSepList Set HTTP headers for all transactions
--http-no-head Don't use HEAD requests to find file sizes in dir listing
--http-no-slash Set this if the site doesn't end directories with /
--http-url string URL of http host to connect to
--hubic-auth-url string Auth server URL.
--hubic-chunk-size SizeSuffix Above this size files will be chunked into a _segments container. (default 5G)
--hubic-client-id string OAuth Client Id
--hubic-client-secret string OAuth Client Secret
--hubic-encoding MultiEncoder This sets the encoding for the backend. (default Slash,InvalidUtf8)
--hubic-no-chunk Don't chunk files during streaming upload.
--hubic-token string OAuth Access Token as a JSON blob.
--hubic-token-url string Token server url.
--jottacloud-encoding MultiEncoder This sets the encoding for the backend. (default Slash,LtGt,DoubleQuote,Colon,Question,Asterisk,Pipe,Del,Ctl,InvalidUtf8,Dot)
--jottacloud-hard-delete Delete files permanently rather than putting them into the trash.
--jottacloud-md5-memory-limit SizeSuffix Files bigger than this will be cached on disk to calculate the MD5 if required. (default 10M)
--jottacloud-trashed-only Only show files that are in the trash.
--jottacloud-upload-resume-limit SizeSuffix Files bigger than this can be resumed if the upload fail's. (default 10M)
--koofr-encoding MultiEncoder This sets the encoding for the backend. (default Slash,BackSlash,Del,Ctl,InvalidUtf8,Dot)
--koofr-endpoint string The Koofr API endpoint to use (default "https://app.koofr.net")
--koofr-mountid string Mount ID of the mount to use. If omitted, the primary mount is used.
--koofr-password string Your Koofr password for rclone (generate one at https://app.koofr.net/app/admin/preferences/password) (obscured)
--koofr-setmtime Does the backend support setting modification time. Set this to false if you use a mount ID that points to a Dropbox or Amazon Drive backend. (default true)
--koofr-user string Your Koofr user name
-l, --links Translate symlinks to/from regular files with a '.rclonelink' extension
--local-case-insensitive Force the filesystem to report itself as case insensitive
--local-case-sensitive Force the filesystem to report itself as case sensitive.
--local-encoding MultiEncoder This sets the encoding for the backend. (default Slash,Dot)
--local-no-check-updated Don't check to see if the files change during upload
--local-no-sparse Disable sparse files for multi-thread downloads
--local-no-unicode-normalization Don't apply unicode normalization to paths and filenames (Deprecated)
--local-nounc string Disable UNC (long path names) conversion on Windows
--local-zero-size-links Assume the Stat size of links is zero (and read them instead)
--mailru-check-hash What should copy do if file checksum is mismatched or invalid (default true)
--mailru-encoding MultiEncoder This sets the encoding for the backend. (default Slash,LtGt,DoubleQuote,Colon,Question,Asterisk,Pipe,BackSlash,Del,Ctl,InvalidUtf8,Dot)
--mailru-pass string Password (obscured)
--mailru-speedup-enable Skip full upload if there is another file with same data hash. (default true)
--mailru-speedup-file-patterns string Comma separated list of file name patterns eligible for speedup (put by hash). (default "*.mkv,*.avi,*.mp4,*.mp3,*.zip,*.gz,*.rar,*.pdf")
--mailru-speedup-max-disk SizeSuffix This option allows you to disable speedup (put by hash) for large files (default 3G)
--mailru-speedup-max-memory SizeSuffix Files larger than the size given below will always be hashed on disk. (default 32M)
--mailru-user string User name (usually email)
--mega-debug Output more debug from Mega.
--mega-encoding MultiEncoder This sets the encoding for the backend. (default Slash,InvalidUtf8,Dot)
--mega-hard-delete Delete files permanently rather than putting them into the trash.
--onedrive-drive-id string The ID of the drive to use
--onedrive-drive-type string The type of the drive ( personal | business | documentLibrary )
--onedrive-encoding MultiEncoder This sets the encoding for the backend. (default Slash,LtGt,DoubleQuote,Colon,Question,Asterisk,Pipe,BackSlash,Del,Ctl,LeftSpace,LeftTilde,RightSpace,RightPeriod,InvalidUtf8,Dot)
--onedrive-expose-onenote-files Set to make OneNote files show up in directory listings.
--onedrive-link-password string Set the password for links created by the link command.
--onedrive-link-scope string Set the scope of the links created by the link command. (default "anonymous")
--onedrive-link-type string Set the type of the links created by the link command. (default "view")
--onedrive-no-versions Remove all versions on modifying operations
--onedrive-region string Choose national cloud region for OneDrive. (default "global")
--onedrive-server-side-across-configs Allow server-side operations (e.g. copy) to work across different onedrive configs.
--onedrive-token string OAuth Access Token as a JSON blob.
--onedrive-token-url string Token server url.
--opendrive-chunk-size SizeSuffix Files will be uploaded in chunks this size. (default 10M)
--opendrive-encoding MultiEncoder This sets the encoding for the backend. (default Slash,LtGt,DoubleQuote,Colon,Question,Asterisk,Pipe,BackSlash,LeftSpace,LeftCrLfHtVt,RightSpace,RightCrLfHtVt,InvalidUtf8,Dot)
--opendrive-password string Password. (obscured)
--opendrive-username string Username
--pcloud-auth-url string Auth server URL.
--pcloud-client-id string OAuth Client Id
--pcloud-client-secret string OAuth Client Secret
--pcloud-encoding MultiEncoder This sets the encoding for the backend. (default Slash,BackSlash,Del,Ctl,InvalidUtf8,Dot)
--pcloud-hostname string Hostname to connect to. (default "api.pcloud.com")
--pcloud-root-folder-id string Fill in for rclone to use a non root folder as its starting point. (default "d0")
--pcloud-token string OAuth Access Token as a JSON blob.
--pcloud-token-url string Token server url.
--premiumizeme-encoding MultiEncoder This sets the encoding for the backend. (default Slash,DoubleQuote,BackSlash,Del,Ctl,InvalidUtf8,Dot)
--putio-encoding MultiEncoder This sets the encoding for the backend. (default Slash,BackSlash,Del,Ctl,InvalidUtf8,Dot)
--qingstor-access-key-id string QingStor Access Key ID
--qingstor-chunk-size SizeSuffix Chunk size to use for uploading. (default 4M)
--qingstor-connection-retries int Number of connection retries. (default 3)
--qingstor-encoding MultiEncoder This sets the encoding for the backend. (default Slash,Ctl,InvalidUtf8)
--qingstor-endpoint string Enter an endpoint URL to connection QingStor API.
--qingstor-env-auth Get QingStor credentials from runtime. Only applies if access_key_id and secret_access_key is blank.
--sftp-key-file string Path to PEM-encoded private key file, leave blank or set key-use-agent to use ssh-agent.
--sftp-key-file-pass string The passphrase to decrypt the PEM-encoded private key file. (obscured)
--sftp-key-pem string Raw PEM-encoded private key, If specified, will override key_file parameter.
--sftp-key-use-agent When set forces the usage of the ssh-agent.
--sftp-known-hosts-file string Optional path to known_hosts file.
--sftp-md5sum-command string The command used to read md5 hashes. Leave blank for autodetect.
--sftp-pass string SSH password, leave blank to use ssh-agent. (obscured)
--sftp-path-override string Override path used by SSH connection.
--sftp-port string SSH port, leave blank to use default (22)
--sftp-pubkey-file string Optional path to public key file.
--sftp-server-command string Specifies the path or command to run a sftp server on the remote host.
--sftp-set-modtime Set the modified time on the remote if set. (default true)
--sftp-sha1sum-command string The command used to read sha1 hashes. Leave blank for autodetect.
--sftp-skip-links Set to skip any symlinks and any other non regular files.
--sftp-subsystem string Specifies the SSH2 subsystem on the remote host. (default "sftp")
--sftp-use-fstat If set use fstat instead of stat
--sftp-use-insecure-cipher Enable the use of insecure ciphers and key exchange methods.
--sftp-user string SSH username, leave blank for current username, $USER
--sharefile-chunk-size SizeSuffix Upload chunk size. Must a power of 2 >= 256k. (default 64M)
--sharefile-encoding MultiEncoder This sets the encoding for the backend. (default Slash,LtGt,DoubleQuote,Colon,Question,Asterisk,Pipe,BackSlash,Ctl,LeftSpace,LeftPeriod,RightSpace,RightPeriod,InvalidUtf8,Dot)
--sharefile-endpoint string Endpoint for API calls.
--sharefile-root-folder-id string ID of the root folder
--sharefile-upload-cutoff SizeSuffix Cutoff for switching to multipart upload. (default 128M)
--skip-links Don't warn about skipped symlinks.
--swift-tenant-id string Tenant ID - optional for v1 auth, this or tenant required otherwise (OS_TENANT_ID)
--swift-user string User name to log in (OS_USERNAME).
--swift-user-id string User ID to log in - optional - most swift systems use user and leave this blank (v3 auth) (OS_USER_ID).
--tardigrade-access-grant string Access Grant.
--tardigrade-api-key string API Key.
--tardigrade-passphrase string Encryption Passphrase. To access existing objects enter passphrase used for uploading.
--tardigrade-provider string Choose an authentication method. (default "existing")
--tardigrade-satellite-address <nodeid>@<address>:<port> Satellite Address. Custom satellite address should match the format: <nodeid>@<address>:<port>. (default "us-central-1.tardigrade.io")
--union-action-policy string Policy to choose upstream on ACTION category. (default "epall")
--union-cache-time int Cache time of usage and free space (in seconds). This option is only useful when a path preserving policy is used. (default 120)
--union-create-policy string Policy to choose upstream on CREATE category. (default "epmfs")
--union-search-policy string Policy to choose upstream on SEARCH category. (default "ff")
--union-upstreams string List of space separated upstreams.
--webdav-bearer-token string Bearer token instead of user/pass (e.g. a Macaroon)
--webdav-bearer-token-command string Command to run to get a bearer token
<p>This is a backend for the <ahref="https://1fichier.com">1fichier</a> cloud storage service. Note that a Premium subscription is required to use the API.</p>
<p>Paths are specified as <code>remote:path</code></p>
<p>Paths may be as deep as required, e.g. <code>remote:directory/subdirectory</code>.</p>
<p>The initial setup for 1Fichier involves getting the API key from the website which you need to do in your browser.</p>
<p>Here is an example of how to make a remote called <code>remote</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> remote
Type of storage to configure.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
[snip]
XX / 1Fichier
\ "fichier"
[snip]
Storage> fichier
** See help for fichier backend at: https://rclone.org/fichier/ **
<p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a> the following characters are also replaced:</p>
<table>
<thead>
<trclass="header">
<th>Character</th>
<thstyle="text-align: center;">Value</th>
<thstyle="text-align: center;">Replacement</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>\</td>
<tdstyle="text-align: center;">0x5C</td>
<tdstyle="text-align: center;">＼</td>
</tr>
<trclass="even">
<td><</td>
<tdstyle="text-align: center;">0x3C</td>
<tdstyle="text-align: center;">＜</td>
</tr>
<trclass="odd">
<td>></td>
<tdstyle="text-align: center;">0x3E</td>
<tdstyle="text-align: center;">＞</td>
</tr>
<trclass="even">
<td>"</td>
<tdstyle="text-align: center;">0x22</td>
<tdstyle="text-align: center;">＂</td>
</tr>
<trclass="odd">
<td>$</td>
<tdstyle="text-align: center;">0x24</td>
<tdstyle="text-align: center;">＄</td>
</tr>
<trclass="even">
<td>`</td>
<tdstyle="text-align: center;">0x60</td>
<tdstyle="text-align: center;">｀</td>
</tr>
<trclass="odd">
<td>'</td>
<tdstyle="text-align: center;">0x27</td>
<tdstyle="text-align: center;">＇</td>
</tr>
</tbody>
</table>
<p>File names can also not start or end with the following characters. These only get replaced if they are the first or last character in the name:</p>
<table>
<thead>
<trclass="header">
<th>Character</th>
<thstyle="text-align: center;">Value</th>
<thstyle="text-align: center;">Replacement</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>SP</td>
<tdstyle="text-align: center;">0x20</td>
<tdstyle="text-align: center;">␠</td>
</tr>
</tbody>
</table>
<p>Invalid UTF-8 bytes will also be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as they can't be used in JSON strings.</p>
<h3id="standard-options">Standard Options</h3>
<p>Here are the standard options specific to fichier (1Fichier).</p>
<h4id="fichier-api-key">--fichier-api-key</h4>
<p>Your API Key, get it from https://1fichier.com/console/params.pl</p>
<ul>
<li>Config: api_key</li>
<li>Env Var: RCLONE_FICHIER_API_KEY</li>
<li>Type: string</li>
<li>Default: ""</li>
</ul>
<h3id="advanced-options">Advanced Options</h3>
<p>Here are the advanced options specific to fichier (1Fichier).</p>
<p><code>rclone about</code> is not supported by the 1Fichier backend. Backends without this capability cannot determine free space for an rclone mount or use policy <code>mfs</code> (most free space) as a member of an rclone union remote.</p>
<p>See <ahref="https://rclone.org/overview/#optional-features">List of backends that do not support rclone about</a> See <ahref="https://rclone.org/commands/rclone_about/">rclone about</a></p>
<h2id="alias">Alias</h2>
<p>The <code>alias</code> remote provides a new name for another remote.</p>
<p>Paths may be as deep as required or a local path, e.g. <code>remote:directory/subdirectory</code> or <code>/directory/subdirectory</code>.</p>
<p>During the initial setup with <code>rclone config</code> you will specify the target remote. The target remote can either be a local path or another remote.</p>
<p>Subfolders can be used in target remote. Assume an alias remote named <code>backup</code> with the target <code>mydrive:private/backup</code>. Invoking <code>rclone mkdir backup:desktop</code> is exactly the same as invoking <code>rclone mkdir mydrive:private/backup/desktop</code>.</p>
<p>There will be no special handling of paths containing <code>..</code> segments. Invoking <code>rclone mkdir backup:../desktop</code> is exactly the same as invoking <code>rclone mkdir mydrive:private/backup/../desktop</code>. The empty path is not allowed as a remote. To alias the current directory use <code>.</code> instead.</p>
<p>Here is an example of how to make an alias called <code>remote</code> for local folder. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> remote
Type of storage to configure.
Choose a number from below, or type in your own value
[snip]
XX / Alias for an existing remote
\ "alias"
[snip]
Storage> alias
Remote or path to alias.
Can be "myremote:path/to/dir", "myremote:bucket", "myremote:" or "/local/path".
<p>Here are the standard options specific to alias (Alias for an existing remote).</p>
<h4id="alias-remote">--alias-remote</h4>
<p>Remote or path to alias. Can be "myremote:path/to/dir", "myremote:bucket", "myremote:" or "/local/path".</p>
<ul>
<li>Config: remote</li>
<li>Env Var: RCLONE_ALIAS_REMOTE</li>
<li>Type: string</li>
<li>Default: ""</li>
</ul>
<h2id="amazon-drive">Amazon Drive</h2>
<p>Amazon Drive, formerly known as Amazon Cloud Drive, is a cloud storage service run by Amazon for consumers.</p>
<h2id="status">Status</h2>
<p><strong>Important:</strong> rclone supports Amazon Drive only if you have your own set of API keys. Unfortunately the <ahref="https://developer.amazon.com/amazon-drive">Amazon Drive developer program</a> is now closed to new entries so if you don't already have your own set of keys you will not be able to use rclone with Amazon Drive.</p>
<p>For the history on why rclone no longer has a set of Amazon Drive API keys see <ahref="https://forum.rclone.org/t/rclone-has-been-banned-from-amazon-drive/2314">the forum</a>.</p>
<p>If you happen to know anyone who works at Amazon then please ask them to re-instate rclone into the Amazon Drive developer program - thanks!</p>
<h2id="setup">Setup</h2>
<p>The initial setup for Amazon Drive involves getting a token from Amazon which you need to do in your browser. <code>rclone config</code> walks you through it.</p>
<p>The configuration process for Amazon Drive may involve using an <ahref="https://github.com/ncw/oauthproxy">oauth proxy</a>. This is used to keep the Amazon credentials out of the source code. The proxy runs in Google's very secure App Engine environment and doesn't store any credentials which pass through it.</p>
<p>Since rclone doesn't currently have its own Amazon Drive credentials so you will either need to have your own <code>client_id</code> and <code>client_secret</code> with Amazon Drive, or use a third party oauth proxy in which case you will need to enter <code>client_id</code>, <code>client_secret</code>, <code>auth_url</code> and <code>token_url</code>.</p>
<p>Note also if you are not using Amazon's <code>auth_url</code> and <code>token_url</code>, (ie you filled in something for those) then if setting up on a remote machine you can only use the <ahref="https://rclone.org/remote_setup/#configuring-by-copying-the-config-file">copying the config method of configuration</a> - <code>rclone authorize</code> will not work.</p>
<p>Here is an example of how to make a remote called <code>remote</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>No remotes found - make a new one
n) New remote
r) Rename remote
c) Copy remote
s) Set configuration password
q) Quit config
n/r/c/s/q> n
name> remote
Type of storage to configure.
Choose a number from below, or type in your own value
[snip]
XX / Amazon Drive
\ "amazon cloud drive"
[snip]
Storage> amazon cloud drive
Amazon Application Client Id - required.
client_id> your client ID goes here
Amazon Application Client Secret - required.
client_secret> your client secret goes here
Auth server URL - leave blank to use Amazon's.
auth_url> Optional auth URL
Token server url - leave blank to use Amazon's.
token_url> Optional token URL
Remote config
Make sure your Redirect URL is set to "http://127.0.0.1:53682/" in your custom config.
Use auto config?
* Say Y if not sure
* Say N if you are working on a remote or headless machine
y) Yes
n) No
y/n> y
If your browser doesn't open automatically go to the following link: http://127.0.0.1:53682/auth
<p>See the <ahref="https://rclone.org/remote_setup/">remote setup docs</a> for how to set it up on a machine with no Internet browser available.</p>
<p>Note that rclone runs a webserver on your local machine to collect the token as returned from Amazon. This only runs from the moment it opens your browser to the moment you get back the verification code. This is on <code>http://127.0.0.1:53682/</code> and this it may require you to unblock it temporarily if you are running a host firewall.</p>
<p>Once configured you can then use <code>rclone</code> like this,</p>
<p>List directories in top level of your Amazon Drive</p>
<pre><code>rclone lsd remote:</code></pre>
<p>List all the files in your Amazon Drive</p>
<pre><code>rclone ls remote:</code></pre>
<p>To copy a local directory to an Amazon Drive directory called backup</p>
<p>Invalid UTF-8 bytes will also be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as they can't be used in JSON strings.</p>
<h3id="deleting-files">Deleting files</h3>
<p>Any files you delete with rclone will end up in the trash. Amazon don't provide an API to permanently delete files, nor to empty the trash, so you will have to do that with one of Amazon's apps or via the Amazon Drive website. As of November 17, 2016, files are automatically deleted by Amazon from the trash after 30 days.</p>
<h3id="using-with-non-.com-amazon-accounts">Using with non <code>.com</code> Amazon accounts</h3>
<p>Let's say you usually use <code>amazon.co.uk</code>. When you authenticate with rclone it will take you to an <code>amazon.com</code> page to log in. Your <code>amazon.co.uk</code> email and password should work here just fine.</p>
<h3id="standard-options-2">Standard Options</h3>
<p>Here are the standard options specific to amazon cloud drive (Amazon Drive).</p>
<p>Additional time per GB to wait after a failed complete upload to see if it appears.</p>
<p>Sometimes Amazon Drive gives an error when a file has been fully uploaded but the file appears anyway after a little while. This happens sometimes for files over 1GB in size and nearly every time for files bigger than 10GB. This parameter controls the time rclone waits for the file to appear.</p>
<p>The default value for this parameter is 3 minutes per GB, so by default it will wait 3 minutes for every GB uploaded to see if the file appears.</p>
<p>You can disable this feature by setting it to 0. This may cause conflict errors as rclone retries the failed upload but the file will most likely appear correctly eventually.</p>
<p>These values were determined empirically by observing lots of uploads of big files for a range of file sizes.</p>
<p>Upload with the "-v" flag to see more info about what rclone is doing in this situation.</p>
<p>Files >= this size will be downloaded via their tempLink.</p>
<p>Files this size or more will be downloaded via their "tempLink". This is to work around a problem with Amazon Drive which blocks downloads of files bigger than about 10GB. The default for this is 9GB which shouldn't need to be changed.</p>
<p>To download files above this threshold, rclone requests a "tempLink" which downloads the file through a temporary URL directly from the underlying S3 storage.</p>
<ul>
<li>Config: templink_threshold</li>
<li>Env Var: RCLONE_ACD_TEMPLINK_THRESHOLD</li>
<li>Type: SizeSuffix</li>
<li>Default: 9G</li>
</ul>
<h4id="acd-encoding">--acd-encoding</h4>
<p>This sets the encoding for the backend.</p>
<p>See: the <ahref="https://rclone.org/overview/#encoding">encoding section in the overview</a> for more info.</p>
<ul>
<li>Config: encoding</li>
<li>Env Var: RCLONE_ACD_ENCODING</li>
<li>Type: MultiEncoder</li>
<li>Default: Slash,InvalidUtf8,Dot</li>
</ul>
<h3id="limitations-3">Limitations</h3>
<p>Note that Amazon Drive is case insensitive so you can't have a file called "Hello.doc" and one called "hello.doc".</p>
<p>Amazon Drive has rate limiting so you may notice errors in the sync (429 errors). rclone will automatically retry the sync up to 3 times by default (see <code>--retries</code> flag) which should hopefully work around this problem.</p>
<p>Amazon Drive has an internal limit of file sizes that can be uploaded to the service. This limit is not officially published, but all files larger than this will fail.</p>
<p>At the time of writing (Jan 2016) is in the area of 50GB per file. This means that larger files are likely to fail.</p>
<p>Unfortunately there is no way for rclone to see that this failure is because of file size, so it will retry the operation, as any other failure. To avoid this problem, use <code>--max-size 50000M</code> option to limit the maximum size of uploaded files. Note that <code>--max-size</code> does not split files into segments, it only ignores files over this size.</p>
<p><code>rclone about</code> is not supported by the Amazon Drive backend. Backends without this capability cannot determine free space for an rclone mount or use policy <code>mfs</code> (most free space) as a member of an rclone union remote.</p>
<p>See <ahref="https://rclone.org/overview/#optional-features">List of backends that do not support rclone about</a> See <ahref="https://rclone.org/commands/rclone_about/">rclone about</a></p>
<p>The S3 backend can be used with a number of different providers:</p>
<ul>
<li>AWS S3</li>
<li>Alibaba Cloud (Aliyun) Object Storage System (OSS)</li>
<li>Ceph</li>
<li>DigitalOcean Spaces</li>
<li>Dreamhost</li>
<li>IBM COS S3</li>
<li>Minio</li>
<li>Scaleway</li>
<li>StackPath</li>
<li>Tencent Cloud Object Storage (COS)</li>
<li>Wasabi</li>
</ul>
<p>Paths are specified as <code>remote:bucket</code> (or <code>remote:</code> for the <code>lsd</code> command.) You may put subdirectories in too, e.g. <code>remote:bucket/path/to/dir</code>.</p>
<p>Once you have made a remote (see the provider specific section above) you can use it like this:</p>
<p>Here is an example of making an s3 configuration. First run</p>
<pre><code>rclone config</code></pre>
<p>This will guide you through an interactive setup process.</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> remote
Type of storage to configure.
Choose a number from below, or type in your own value
[snip]
XX / Amazon S3 Compliant Storage Providers including AWS, Ceph, Dreamhost, IBM COS, Minio, and Tencent COS
\ "s3"
[snip]
Storage> s3
Choose your S3 provider.
Choose a number from below, or type in your own value
1 / Amazon Web Services (AWS) S3
\ "AWS"
2 / Ceph Object Storage
\ "Ceph"
3 / Digital Ocean Spaces
\ "DigitalOcean"
4 / Dreamhost DreamObjects
\ "Dreamhost"
5 / IBM COS S3
\ "IBMCOS"
6 / Minio Object Storage
\ "Minio"
7 / Wasabi Object Storage
\ "Wasabi"
8 / Any other S3 compatible provider
\ "Other"
provider> 1
Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars). Only applies if access_key_id and secret_access_key is blank.
Choose a number from below, or type in your own value
1 / Enter AWS credentials in the next step
\ "false"
2 / Get AWS credentials from the environment (env vars or IAM)
\ "true"
env_auth> 1
AWS Access Key ID - leave blank for anonymous access or runtime credentials.
access_key_id> XXX
AWS Secret Access Key (password) - leave blank for anonymous access or runtime credentials.
secret_access_key> YYY
Region to connect to.
Choose a number from below, or type in your own value
/ The default endpoint - a good choice if you are unsure.
1 | US Region, Northern Virginia, or Pacific Northwest.
| Leave location constraint empty.
\ "us-east-1"
/ US East (Ohio) Region
2 | Needs location constraint us-east-2.
\ "us-east-2"
/ US West (Oregon) Region
3 | Needs location constraint us-west-2.
\ "us-west-2"
/ US West (Northern California) Region
4 | Needs location constraint us-west-1.
\ "us-west-1"
/ Canada (Central) Region
5 | Needs location constraint ca-central-1.
\ "ca-central-1"
/ EU (Ireland) Region
6 | Needs location constraint EU or eu-west-1.
\ "eu-west-1"
/ EU (London) Region
7 | Needs location constraint eu-west-2.
\ "eu-west-2"
/ EU (Frankfurt) Region
8 | Needs location constraint eu-central-1.
\ "eu-central-1"
/ Asia Pacific (Singapore) Region
9 | Needs location constraint ap-southeast-1.
\ "ap-southeast-1"
/ Asia Pacific (Sydney) Region
10 | Needs location constraint ap-southeast-2.
\ "ap-southeast-2"
/ Asia Pacific (Tokyo) Region
11 | Needs location constraint ap-northeast-1.
\ "ap-northeast-1"
/ Asia Pacific (Seoul)
12 | Needs location constraint ap-northeast-2.
\ "ap-northeast-2"
/ Asia Pacific (Mumbai)
13 | Needs location constraint ap-south-1.
\ "ap-south-1"
/ Asia Pacific (Hong Kong) Region
14 | Needs location constraint ap-east-1.
\ "ap-east-1"
/ South America (Sao Paulo) Region
15 | Needs location constraint sa-east-1.
\ "sa-east-1"
region> 1
Endpoint for S3 API.
Leave blank if using AWS to use the default endpoint for the region.
endpoint>
Location constraint - must be set to match the Region. Used when creating buckets only.
Choose a number from below, or type in your own value
1 / Empty for US Region, Northern Virginia, or Pacific Northwest.
\ ""
2 / US East (Ohio) Region.
\ "us-east-2"
3 / US West (Oregon) Region.
\ "us-west-2"
4 / US West (Northern California) Region.
\ "us-west-1"
5 / Canada (Central) Region.
\ "ca-central-1"
6 / EU (Ireland) Region.
\ "eu-west-1"
7 / EU (London) Region.
\ "eu-west-2"
8 / EU Region.
\ "EU"
9 / Asia Pacific (Singapore) Region.
\ "ap-southeast-1"
10 / Asia Pacific (Sydney) Region.
\ "ap-southeast-2"
11 / Asia Pacific (Tokyo) Region.
\ "ap-northeast-1"
12 / Asia Pacific (Seoul)
\ "ap-northeast-2"
13 / Asia Pacific (Mumbai)
\ "ap-south-1"
14 / Asia Pacific (Hong Kong)
\ "ap-east-1"
15 / South America (Sao Paulo) Region.
\ "sa-east-1"
location_constraint> 1
Canned ACL used when creating buckets and/or storing objects in S3.
For more info visit https://docs.aws.amazon.com/AmazonS3/latest/dev/acl-overview.html#canned-acl
Choose a number from below, or type in your own value
1 / Owner gets FULL_CONTROL. No one else has access rights (default).
\ "private"
2 / Owner gets FULL_CONTROL. The AllUsers group gets READ access.
\ "public-read"
/ Owner gets FULL_CONTROL. The AllUsers group gets READ and WRITE access.
3 | Granting this on a bucket is generally not recommended.
\ "public-read-write"
4 / Owner gets FULL_CONTROL. The AuthenticatedUsers group gets READ access.
5 | If you specify this canned ACL when creating a bucket, Amazon S3 ignores it.
\ "bucket-owner-read"
/ Both the object owner and the bucket owner get FULL_CONTROL over the object.
6 | If you specify this canned ACL when creating a bucket, Amazon S3 ignores it.
\ "bucket-owner-full-control"
acl> 1
The server-side encryption algorithm used when storing this object in S3.
Choose a number from below, or type in your own value
1 / None
\ ""
2 / AES256
\ "AES256"
server_side_encryption> 1
The storage class to use when storing objects in S3.
Choose a number from below, or type in your own value
1 / Default
\ ""
2 / Standard storage class
\ "STANDARD"
3 / Reduced redundancy storage class
\ "REDUCED_REDUNDANCY"
4 / Standard Infrequent Access storage class
\ "STANDARD_IA"
5 / One Zone Infrequent Access storage class
\ "ONEZONE_IA"
6 / Glacier storage class
\ "GLACIER"
7 / Glacier Deep Archive storage class
\ "DEEP_ARCHIVE"
8 / Intelligent-Tiering storage class
\ "INTELLIGENT_TIERING"
storage_class> 1
Remote config
--------------------
[remote]
type = s3
provider = AWS
env_auth = false
access_key_id = XXX
secret_access_key = YYY
region = us-east-1
endpoint =
location_constraint =
acl = private
server_side_encryption =
storage_class =
--------------------
y) Yes this is OK
e) Edit this remote
d) Delete this remote
y/e/d></code></pre>
<h3id="modified-time">Modified time</h3>
<p>The modified time is stored as metadata on the object as <code>X-Amz-Meta-Mtime</code> as floating point since the epoch accurate to 1 ns.</p>
<p>If the modification time needs to be updated rclone will attempt to perform a server side copy to update the modification if the object can be copied in a single part. In the case the object is larger than 5Gb or is in Glacier or Glacier Deep Archive storage the object will be uploaded rather than copied.</p>
<p>Note that reading this from the object takes an additional <code>HEAD</code> request as the metadata isn't returned in object listings.</p>
<h3id="reducing-costs">Reducing costs</h3>
<h4id="avoiding-head-requests-to-read-the-modification-time">Avoiding HEAD requests to read the modification time</h4>
<p>By default rclone will use the modification time of objects stored in S3 for syncing. This is stored in object metadata which unfortunately takes an extra HEAD request to read which can be expensive (in time and money).</p>
<p>The modification time is used by default for all operations that require checking the time a file was last updated. It allows rclone to treat the remote more like a true filesystem, but it is inefficient on S3 because it requires an extra API call to retrieve the metadata.</p>
<p>The extra API calls can be avoided when syncing (using <code>rclone sync</code> or <code>rclone copy</code>) in a few different ways, each with its own tradeoffs.</p>
<ul>
<li><code>--size-only</code>
<ul>
<li>Only checks the size of files.</li>
<li>Uses no extra transactions.</li>
<li>If the file doesn't change size then rclone won't detect it has changed.</li>
<li>Modification time becomes the time the object was uploaded.</li>
<li>For many operations this is sufficient to determine if it needs uploading.</li>
<li>Using <code>--update</code> along with <code>--use-server-modtime</code>, avoids the extra API call and uploads files whose local modification time is newer than the time it was last uploaded.</li>
<li>Files created with timestamps in the past will be missed by the sync.</li>
<p>These flags can and should be used in combination with <code>--fast-list</code> - see below.</p>
<p>If using <code>rclone mount</code> or any command using the VFS (eg <code>rclone serve</code>) commands then you might want to consider using the VFS flag <code>--no-modtime</code> which will stop rclone reading the modification time for every object. You could also use <code>--use-server-modtime</code> if you are happy with the modification times of the objects being the time of upload.</p>
<h4id="avoiding-get-requests-to-read-directory-listings">Avoiding GET requests to read directory listings</h4>
<p>Rclone's default directory traversal is to process each directory individually. This takes one API call per directory. Using the <code>--fast-list</code> flag will read all info about the the objects into memory first using a smaller number of API calls (one per 1000 objects). See the <ahref="https://rclone.org/docs/#fast-list">rclone docs</a> for more details.</p>
<p><code>--fast-list</code> trades off API transactions for memory use. As a rough guide rclone uses 1k of memory per object stored, so using <code>--fast-list</code> on a sync of a million objects will use roughly 1 GB of RAM.</p>
<p>If you are only copying a small number of files into a big repository then using <code>--no-traverse</code> is a good idea. This finds objects directly instead of through directory listings. You can do a "top-up" sync very cheaply by using <code>--max-age</code> and <code>--no-traverse</code> to copy only recent files, eg</p>
<p>For small objects which weren't uploaded as multipart uploads (objects sized below <code>--s3-upload-cutoff</code> if uploaded with rclone) rclone uses the <code>ETag:</code> header as an MD5 checksum.</p>
<p>However for objects which were uploaded as multipart uploads or with server side encryption (SSE-AWS or SSE-C) the <code>ETag</code> header is no longer the MD5 sum of the data, so rclone adds an additional piece of metadata <code>X-Amz-Meta-Md5chksum</code> which is a base64 encoded MD5 hash (in the same format as is required for <code>Content-MD5</code>).</p>
<p>For large objects, calculating this hash can take some time so the addition of this hash can be disabled with <code>--s3-disable-checksum</code>. This will mean that these objects do not have an MD5 checksum.</p>
<p>Note that reading this from the object takes an additional <code>HEAD</code> request as the metadata isn't returned in object listings.</p>
<h3id="cleanup-1">Cleanup</h3>
<p>If you run <code>rclone cleanup s3:bucket</code> then it will remove all pending multipart uploads older than 24 hours. You can use the <code>-i</code> flag to see exactly what it will do. If you want more control over the expiry date then run <code>rclone backend cleanup s3:bucket -o max-age=1h</code> to expire all uploads older than one hour. You can use <code>rclone backend list-multipart-uploads s3:bucket</code> to see the pending multipart uploads.</p>
<p>Invalid UTF-8 bytes will be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as they can't be used in XML.</p>
<p>The following characters are replaced since these are problematic when dealing with the REST API:</p>
<table>
<thead>
<trclass="header">
<th>Character</th>
<thstyle="text-align: center;">Value</th>
<thstyle="text-align: center;">Replacement</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>NUL</td>
<tdstyle="text-align: center;">0x00</td>
<tdstyle="text-align: center;">␀</td>
</tr>
<trclass="even">
<td>/</td>
<tdstyle="text-align: center;">0x2F</td>
<tdstyle="text-align: center;">／</td>
</tr>
</tbody>
</table>
<p>The encoding will also encode these file names as they don't seem to work with the SDK properly:</p>
<table>
<thead>
<trclass="header">
<th>File name</th>
<thstyle="text-align: center;">Replacement</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>.</td>
<tdstyle="text-align: center;">．</td>
</tr>
<trclass="even">
<td>..</td>
<tdstyle="text-align: center;">．．</td>
</tr>
</tbody>
</table>
<h3id="multipart-uploads">Multipart uploads</h3>
<p>rclone supports multipart uploads with S3 which means that it can upload files bigger than 5GB.</p>
<p>Note that files uploaded <em>both</em> with multipart upload <em>and</em> through crypt remotes do not have MD5 sums.</p>
<p>rclone switches from single part uploads to multipart uploads at the point specified by <code>--s3-upload-cutoff</code>. This can be a maximum of 5GB and a minimum of 0 (ie always upload multipart files).</p>
<p>The chunk sizes used in the multipart upload are specified by <code>--s3-chunk-size</code> and the number of chunks uploaded concurrently is specified by <code>--s3-upload-concurrency</code>.</p>
<p>Multipart uploads will use <code>--transfers</code> * <code>--s3-upload-concurrency</code> * <code>--s3-chunk-size</code> extra memory. Single part uploads to not use extra memory.</p>
<p>Single part transfers can be faster than multipart transfers or slower depending on your latency from S3 - the more latency, the more likely single part transfers will be faster.</p>
<p>Increasing <code>--s3-upload-concurrency</code> will increase throughput (8 would be a sensible value) and increasing <code>--s3-chunk-size</code> also increases throughput (16M would be sensible). Increasing either of these will use more memory. The default values are high enough to gain most of the possible performance without using too much memory.</p>
<h3id="buckets-and-regions">Buckets and Regions</h3>
<p>With Amazon S3 you can list buckets (<code>rclone lsd</code>) using any region, but you can only access the content of a bucket from the region it was created in. If you attempt to access a bucket from the wrong region, you will get an error, <code>incorrect region, the bucket is not in 'XXX' region</code>.</p>
<h3id="authentication-4">Authentication</h3>
<p>There are a number of ways to supply <code>rclone</code> with a set of AWS credentials, with and without using the environment.</p>
<p>The different authentication methods are tried in this order:</p>
<ul>
<li>Directly in the rclone configuration file (<code>env_auth = false</code> in the config file):
<ul>
<li><code>access_key_id</code> and <code>secret_access_key</code> are required.</li>
<li><code>session_token</code> can be optionally set when using AWS STS.</li>
</ul></li>
<li>Runtime configuration (<code>env_auth = true</code> in the config file):
<ul>
<li>Export the following environment variables before running <code>rclone</code>:
<ul>
<li>Access Key ID: <code>AWS_ACCESS_KEY_ID</code> or <code>AWS_ACCESS_KEY</code></li>
<li>Secret Access Key: <code>AWS_SECRET_ACCESS_KEY</code> or <code>AWS_SECRET_KEY</code></li>
<li>Or, use a <ahref="https://docs.aws.amazon.com/cli/latest/userguide/cli-multiple-profiles.html">named profile</a>:
<ul>
<li>Profile files are standard files used by AWS CLI tools</li>
<li>By default it will use the profile in your home directory (e.g. <code>~/.aws/credentials</code> on unix based systems) file and the "default" profile, to change set these environment variables:
<ul>
<li><code>AWS_SHARED_CREDENTIALS_FILE</code> to control which file.</li>
<li><code>AWS_PROFILE</code> to control which profile to use.</li>
</ul></li>
</ul></li>
<li>Or, run <code>rclone</code> in an ECS task with an IAM role (AWS only).</li>
<li>Or, run <code>rclone</code> on an EC2 instance with an IAM role (AWS only).</li>
<li>Or, run <code>rclone</code> in an EKS pod with an IAM role that is associated with a service account (AWS only).</li>
</ul></li>
</ul>
<p>If none of these option actually end up providing <code>rclone</code> with AWS credentials then S3 interaction will be non-authenticated (see below).</p>
<h3id="s3-permissions">S3 Permissions</h3>
<p>When using the <code>sync</code> subcommand of <code>rclone</code> the following minimum permissions are required to be available on the bucket being written to:</p>
<ul>
<li><code>ListBucket</code></li>
<li><code>DeleteObject</code></li>
<li><code>GetObject</code></li>
<li><code>PutObject</code></li>
<li><code>PutObjectACL</code></li>
</ul>
<p>When using the <code>lsd</code> subcommand, the <code>ListAllMyBuckets</code> permission is required.</p>
<li>This is a policy that can be used when creating bucket. It assumes that <code>USER_NAME</code> has been created.</li>
<li>The Resource entry must include both resource ARNs, as one implies the bucket and the other implies the bucket's objects.</li>
</ol>
<p>For reference, <ahref="https://gist.github.com/ebridges/ebfc9042dd7c756cd101cfa807b7ae2b">here's an Ansible script</a> that will generate one or more buckets that will work with <code>rclone sync</code>.</p>
<h3id="key-management-system-kms">Key Management System (KMS)</h3>
<p>If you are using server-side encryption with KMS then you must make sure rclone is configured with <code>server_side_encryption = aws:kms</code> otherwise you will find you can't transfer small objects - these will create checksum errors.</p>
<h3id="glacier-and-glacier-deep-archive">Glacier and Glacier Deep Archive</h3>
<p>You can upload objects using the glacier storage class or transition them to glacier using a <ahref="http://docs.aws.amazon.com/AmazonS3/latest/user-guide/create-lifecycle.html">lifecycle policy</a>. The bucket can still be synced or copied into normally, but if rclone tries to access data from the glacier storage class you will see an error like below.</p>
<pre><code>2017/09/11 19:07:43 Failed to sync: failed to open source object: Object in GLACIER, restore first: path/to/file</code></pre>
<p>In this case you need to <ahref="http://docs.aws.amazon.com/AmazonS3/latest/user-guide/restore-archived-objects.html">restore</a> the object(s) in question before using rclone.</p>
<p>Note that rclone only speaks the S3 API it does not speak the Glacier Vault API, so rclone cannot directly access Glacier Vaults.</p>
<h3id="standard-options-3">Standard Options</h3>
<p>Here are the standard options specific to s3 (Amazon S3 Compliant Storage Providers including AWS, Alibaba, Ceph, Digital Ocean, Dreamhost, IBM COS, Minio, and Tencent COS).</p>
<h4id="s3-provider">--s3-provider</h4>
<p>Choose your S3 provider.</p>
<ul>
<li>Config: provider</li>
<li>Env Var: RCLONE_S3_PROVIDER</li>
<li>Type: string</li>
<li>Default: ""</li>
<li>Examples:
<ul>
<li>"AWS"
<ul>
<li>Amazon Web Services (AWS) S3</li>
</ul></li>
<li>"Alibaba"
<ul>
<li>Alibaba Cloud Object Storage System (OSS) formerly Aliyun</li>
</ul></li>
<li>"Ceph"
<ul>
<li>Ceph Object Storage</li>
</ul></li>
<li>"DigitalOcean"
<ul>
<li>Digital Ocean Spaces</li>
</ul></li>
<li>"Dreamhost"
<ul>
<li>Dreamhost DreamObjects</li>
</ul></li>
<li>"IBMCOS"
<ul>
<li>IBM COS S3</li>
</ul></li>
<li>"Minio"
<ul>
<li>Minio Object Storage</li>
</ul></li>
<li>"Netease"
<ul>
<li>Netease Object Storage (NOS)</li>
</ul></li>
<li>"Scaleway"
<ul>
<li>Scaleway Object Storage</li>
</ul></li>
<li>"StackPath"
<ul>
<li>StackPath Object Storage</li>
</ul></li>
<li>"TencentCOS"
<ul>
<li>Tencent Cloud Object Storage (COS)</li>
</ul></li>
<li>"Wasabi"
<ul>
<li>Wasabi Object Storage</li>
</ul></li>
<li>"Other"
<ul>
<li>Any other S3 compatible provider</li>
</ul></li>
</ul></li>
</ul>
<h4id="s3-env-auth">--s3-env-auth</h4>
<p>Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars). Only applies if access_key_id and secret_access_key is blank.</p>
<ul>
<li>Config: env_auth</li>
<li>Env Var: RCLONE_S3_ENV_AUTH</li>
<li>Type: bool</li>
<li>Default: false</li>
<li>Examples:
<ul>
<li>"false"
<ul>
<li>Enter AWS credentials in the next step</li>
</ul></li>
<li>"true"
<ul>
<li>Get AWS credentials from the environment (env vars or IAM)</li>
</ul></li>
</ul></li>
</ul>
<h4id="s3-access-key-id">--s3-access-key-id</h4>
<p>AWS Access Key ID. Leave blank for anonymous access or runtime credentials.</p>
<li>If you specify this canned ACL when creating a bucket, Amazon S3 ignores it.</li>
</ul></li>
<li>"bucket-owner-full-control"
<ul>
<li>Both the object owner and the bucket owner get FULL_CONTROL over the object.</li>
<li>If you specify this canned ACL when creating a bucket, Amazon S3 ignores it.</li>
</ul></li>
<li>"private"
<ul>
<li>Owner gets FULL_CONTROL. No one else has access rights (default). This acl is available on IBM Cloud (Infra), IBM Cloud (Storage), On-Premise COS</li>
</ul></li>
<li>"public-read"
<ul>
<li>Owner gets FULL_CONTROL. The AllUsers group gets READ access. This acl is available on IBM Cloud (Infra), IBM Cloud (Storage), On-Premise IBM COS</li>
</ul></li>
<li>"public-read-write"
<ul>
<li>Owner gets FULL_CONTROL. The AllUsers group gets READ and WRITE access. This acl is available on IBM Cloud (Infra), On-Premise IBM COS</li>
</ul></li>
<li>"authenticated-read"
<ul>
<li>Owner gets FULL_CONTROL. The AuthenticatedUsers group gets READ access. Not supported on Buckets. This acl is available on IBM Cloud (Infra) and On-Premise IBM COS</li>
<p>The storage class to use when storing new objects in S3.</p>
<ul>
<li>Config: storage_class</li>
<li>Env Var: RCLONE_S3_STORAGE_CLASS</li>
<li>Type: string</li>
<li>Default: ""</li>
<li>Examples:
<ul>
<li>""
<ul>
<li>Default</li>
</ul></li>
<li>"STANDARD"
<ul>
<li>The Standard class for any upload; suitable for on-demand content like streaming or CDN.</li>
</ul></li>
<li>"GLACIER"
<ul>
<li>Archived storage; prices are lower, but it needs to be restored first to be accessed.</li>
</ul></li>
</ul></li>
</ul>
<h3id="advanced-options-2">Advanced Options</h3>
<p>Here are the advanced options specific to s3 (Amazon S3 Compliant Storage Providers including AWS, Alibaba, Ceph, Digital Ocean, Dreamhost, IBM COS, Minio, and Tencent COS).</p>
<h4id="s3-bucket-acl">--s3-bucket-acl</h4>
<p>Canned ACL used when creating buckets.</p>
<p>For more info visit https://docs.aws.amazon.com/AmazonS3/latest/dev/acl-overview.html#canned-acl</p>
<p>Note that this ACL is applied when only when creating buckets. If it isn't set then "acl" is used instead.</p>
<ul>
<li>Config: bucket_acl</li>
<li>Env Var: RCLONE_S3_BUCKET_ACL</li>
<li>Type: string</li>
<li>Default: ""</li>
<li>Examples:
<ul>
<li>"private"
<ul>
<li>Owner gets FULL_CONTROL. No one else has access rights (default).</li>
</ul></li>
<li>"public-read"
<ul>
<li>Owner gets FULL_CONTROL. The AllUsers group gets READ access.</li>
</ul></li>
<li>"public-read-write"
<ul>
<li>Owner gets FULL_CONTROL. The AllUsers group gets READ and WRITE access.</li>
<li>Granting this on a bucket is generally not recommended.</li>
</ul></li>
<li>"authenticated-read"
<ul>
<li>Owner gets FULL_CONTROL. The AuthenticatedUsers group gets READ access.</li>
<p>If using SSE-C you may provide the secret encryption key MD5 checksum (optional).</p>
<p>If you leave it blank, this is calculated automatically from the sse_customer_key provided.</p>
<ul>
<li>Config: sse_customer_key_md5</li>
<li>Env Var: RCLONE_S3_SSE_CUSTOMER_KEY_MD5</li>
<li>Type: string</li>
<li>Default: ""</li>
<li>Examples:
<ul>
<li>""
<ul>
<li>None</li>
</ul></li>
</ul></li>
</ul>
<h4id="s3-upload-cutoff">--s3-upload-cutoff</h4>
<p>Cutoff for switching to chunked upload</p>
<p>Any files larger than this will be uploaded in chunks of chunk_size. The minimum is 0 and the maximum is 5GB.</p>
<ul>
<li>Config: upload_cutoff</li>
<li>Env Var: RCLONE_S3_UPLOAD_CUTOFF</li>
<li>Type: SizeSuffix</li>
<li>Default: 200M</li>
</ul>
<h4id="s3-chunk-size">--s3-chunk-size</h4>
<p>Chunk size to use for uploading.</p>
<p>When uploading files larger than upload_cutoff or files with unknown size (e.g. from "rclone rcat" or uploaded with "rclone mount" or google photos or google docs) they will be uploaded as multipart uploads using this chunk size.</p>
<p>Note that "--s3-upload-concurrency" chunks of this size are buffered in memory per transfer.</p>
<p>If you are transferring large files over high-speed links and you have enough memory, then increasing this will speed up the transfers.</p>
<p>Rclone will automatically increase the chunk size when uploading a large file of known size to stay below the 10,000 chunks limit.</p>
<p>Files of unknown size are uploaded with the configured chunk_size. Since the default chunk size is 5MB and there can be at most 10,000 chunks, this means that by default the maximum size of a file you can stream upload is 48GB. If you wish to stream upload larger files then you will need to increase chunk_size.</p>
<p>Don't store MD5 checksum with object metadata</p>
<p>Normally rclone will calculate the MD5 checksum of the input before uploading it so it can add it to metadata on the object. This is great for data integrity checking but can cause long delays for large files to start uploading.</p>
<p>If env_auth = true then rclone can use a shared credentials file.</p>
<p>If this variable is empty rclone will look for the "AWS_SHARED_CREDENTIALS_FILE" env variable. If the env value is empty it will default to the current user's home directory.</p>
<p>This is the number of chunks of the same file that are uploaded concurrently.</p>
<p>If you are uploading small numbers of large files over high-speed links and these uploads do not fully utilize your bandwidth, then increasing this may help to speed up the transfers.</p>
<p>If true use path style access if false use virtual hosted style.</p>
<p>If this is true (the default) then rclone will use path style access, if false then rclone will use virtual path style. See <ahref="https://docs.aws.amazon.com/AmazonS3/latest/dev/UsingBucket.html#access-bucket-intro">the AWS S3 docs</a> for more info.</p>
<p>Some providers (e.g. AWS, Aliyun OSS, Netease COS, or Tencent COS) require this set to false - rclone will do this automatically based on the provider setting.</p>
<ul>
<li>Config: force_path_style</li>
<li>Env Var: RCLONE_S3_FORCE_PATH_STYLE</li>
<li>Type: bool</li>
<li>Default: true</li>
</ul>
<h4id="s3-v2-auth">--s3-v2-auth</h4>
<p>If true use v2 authentication.</p>
<p>If this is false (the default) then rclone will use v4 authentication. If it is set then rclone will use v2 authentication.</p>
<p>Use this only if v4 signatures don't work, e.g. pre Jewel/v10 CEPH.</p>
<p>If true avoid calling abort upload on a failure, leaving all successfully uploaded parts on S3 for manual recovery.</p>
<p>It should be set to true for resuming uploads across different sessions.</p>
<p>WARNING: Storing parts of an incomplete multipart upload counts towards space usage on S3 and will add additional costs if not cleaned up.</p>
<ul>
<li>Config: leave_parts_on_error</li>
<li>Env Var: RCLONE_S3_LEAVE_PARTS_ON_ERROR</li>
<li>Type: bool</li>
<li>Default: false</li>
</ul>
<h4id="s3-list-chunk">--s3-list-chunk</h4>
<p>Size of listing chunk (response list for each ListObject S3 request).</p>
<p>This option is also known as "MaxKeys", "max-items", or "page-size" from the AWS S3 specification. Most services truncate the response list to 1000 objects even if requested more than that. In AWS S3 this is a global maximum and cannot be changed, see <ahref="https://docs.aws.amazon.com/cli/latest/reference/s3/ls.html">AWS S3</a>. In Ceph, this can be increased with the "rgw list buckets max chunk" option.</p>
<p>If set, don't attempt to check the bucket exists or create it</p>
<p>This can be useful when trying to minimise the number of transactions rclone does if you know the bucket exists already.</p>
<p>It can also be needed if the user you are using does not have bucket creation permissions. Before v1.52.0 this would have passed silently due to a bug.</p>
<ul>
<li>Config: no_check_bucket</li>
<li>Env Var: RCLONE_S3_NO_CHECK_BUCKET</li>
<li>Type: bool</li>
<li>Default: false</li>
</ul>
<h4id="s3-no-head">--s3-no-head</h4>
<p>If set, don't HEAD uploaded objects to check integrity</p>
<p>This can be useful when trying to minimise the number of transactions rclone does.</p>
<p>Setting it means that if rclone receives a 200 OK message after uploading an object with PUT then it will assume that it got uploaded properly.</p>
<p>In particular it will assume:</p>
<ul>
<li>the metadata, including modtime, storage class and content type was as uploaded</li>
<li>the size was as uploaded</li>
</ul>
<p>It reads the following items from the response for a single part PUT:</p>
<ul>
<li>the MD5SUM</li>
<li>The uploaded date</li>
</ul>
<p>For multipart uploads these items aren't read.</p>
<p>If an source object of unknown length is uploaded then rclone <strong>will</strong> do a HEAD request.</p>
<p>Setting this flag increases the chance for undetected upload failures, in particular an incorrect size, so it isn't recommended for normal operation. In practice the chance of an undetected upload failure is very small even with this flag.</p>
<ul>
<li>Config: no_head</li>
<li>Env Var: RCLONE_S3_NO_HEAD</li>
<li>Type: bool</li>
<li>Default: false</li>
</ul>
<h4id="s3-encoding">--s3-encoding</h4>
<p>This sets the encoding for the backend.</p>
<p>See: the <ahref="https://rclone.org/overview/#encoding">encoding section in the overview</a> for more info.</p>
<p>How often internal memory buffer pools will be flushed. Uploads which requires additional buffers (f.e multipart) will use memory pool for allocations. This option controls how often unused buffers will be removed from the pool.</p>
<p>Whether to use mmap buffers in internal memory pool.</p>
<ul>
<li>Config: memory_pool_use_mmap</li>
<li>Env Var: RCLONE_S3_MEMORY_POOL_USE_MMAP</li>
<li>Type: bool</li>
<li>Default: false</li>
</ul>
<h4id="s3-disable-http2">--s3-disable-http2</h4>
<p>Disable usage of http2 for S3 backends</p>
<p>There is currently an unsolved issue with the s3 (specifically minio) backend and HTTP/2. HTTP/2 is enabled by default for the s3 backend but can be disabled here. When the issue is solved this flag will be removed.</p>
<p>Durations are parsed as per the rest of rclone, 2h, 7d, 7w etc.</p>
<p>Options:</p>
<ul>
<li>"max-age": Max age of upload to delete</li>
</ul>
<h3id="anonymous-access-to-public-buckets">Anonymous access to public buckets</h3>
<p>If you want to use rclone to access a public bucket, configure with a blank <code>access_key_id</code> and <code>secret_access_key</code>. Your config should end up looking like this:</p>
<pre><code>[anons3]
type = s3
provider = AWS
env_auth = false
access_key_id =
secret_access_key =
region = us-east-1
endpoint =
location_constraint =
acl = private
server_side_encryption =
storage_class = </code></pre>
<p>Then use it as normal with the name of the public bucket, e.g.</p>
<p>You will be able to list and copy data but not upload it.</p>
<h3id="ceph">Ceph</h3>
<p><ahref="https://ceph.com/">Ceph</a> is an open source unified, distributed storage system designed for excellent performance, reliability and scalability. It has an S3 compatible object storage interface.</p>
<p>To use rclone with Ceph, configure as above but leave the region blank and set the endpoint. You should end up with something like this in your config:</p>
<pre><code>[ceph]
type = s3
provider = Ceph
env_auth = false
access_key_id = XXX
secret_access_key = YYY
region =
endpoint = https://ceph.endpoint.example.com
location_constraint =
acl =
server_side_encryption =
storage_class =</code></pre>
<p>If you are using an older version of CEPH, e.g. 10.2.x Jewel, then you may need to supply the parameter <code>--s3-upload-cutoff 0</code> or put this in the config file as <code>upload_cutoff 0</code> to work around a bug which causes uploading of small files to fail.</p>
<p>Note also that Ceph sometimes puts <code>/</code> in the passwords it gives users. If you read the secret access key using the command line tools you will get a JSON blob with the <code>/</code> escaped as <code>\/</code>. Make sure you only write <code>/</code> in the secret access key.</p>
<p>Eg the dump from Ceph looks something like this (irrelevant keys removed).</p>
<pre><code>{
"user_id": "xxx",
"display_name": "xxxx",
"keys": [
{
"user": "xxx",
"access_key": "xxxxxx",
"secret_key": "xxxxxx\/xxxx"
}
],
}</code></pre>
<p>Because this is a json dump, it is encoding the <code>/</code> as <code>\/</code>, so if you use the secret key as <code>xxxxxx/xxxx</code> it will work fine.</p>
<h3id="dreamhost">Dreamhost</h3>
<p>Dreamhost <ahref="https://www.dreamhost.com/cloud/storage/">DreamObjects</a> is an object storage system based on CEPH.</p>
<p>To use rclone with Dreamhost, configure as above but leave the region blank and set the endpoint. You should end up with something like this in your config:</p>
<p><ahref="https://www.digitalocean.com/products/object-storage/">Spaces</a> is an <ahref="https://developers.digitalocean.com/documentation/spaces/">S3-interoperable</a> object storage service from cloud provider DigitalOcean.</p>
<p>To connect to DigitalOcean Spaces you will need an access key and secret key. These can be retrieved on the "<ahref="https://cloud.digitalocean.com/settings/api/tokens">Applications & API</a>" page of the DigitalOcean control panel. They will be needed when prompted by <code>rclone config</code> for your <code>access_key_id</code> and <code>secret_access_key</code>.</p>
<p>When prompted for a <code>region</code> or <code>location_constraint</code>, press enter to use the default value. The region must be included in the <code>endpoint</code> setting (e.g. <code>nyc3.digitaloceanspaces.com</code>). The default values can be used for other settings.</p>
<p>Going through the whole process of creating a new remote by running <code>rclone config</code>, each prompt should be answered as shown below:</p>
<pre><code>Storage> s3
env_auth> 1
access_key_id> YOUR_ACCESS_KEY
secret_access_key> YOUR_SECRET_KEY
region>
endpoint> nyc3.digitaloceanspaces.com
location_constraint>
acl>
storage_class></code></pre>
<p>The resulting configuration file should look like:</p>
<pre><code>[spaces]
type = s3
provider = DigitalOcean
env_auth = false
access_key_id = YOUR_ACCESS_KEY
secret_access_key = YOUR_SECRET_KEY
region =
endpoint = nyc3.digitaloceanspaces.com
location_constraint =
acl =
server_side_encryption =
storage_class =</code></pre>
<p>Once configured, you can create a new Space and begin copying files. For example:</p>
<p>Information stored with IBM Cloud Object Storage is encrypted and dispersed across multiple geographic locations, and accessed through an implementation of the S3 API. This service makes use of the distributed storage technologies provided by IBM’s Cloud Object Storage System (formerly Cleversafe). For more information visit: (http://www.ibm.com/cloud/object-storage)</p>
<p>To configure access to IBM COS S3, follow the steps below:</p>
<oltype="1">
<li>Run rclone config and select n for a new remote.</li>
</ol>
<pre><code> 2018/02/14 14:13:11 NOTICE: Config file "C:\\Users\\a\\.config\\rclone\\rclone.conf" not found - using defaults
<li>Specify a IBM COS Location Constraint. The location constraint must match endpoint when using IBM Cloud Public. For on-prem COS, do not make a selection from this list, hit enter</li>
</ol>
<pre><code> 1 / US Cross Region Standard
\ "us-standard"
2 / US Cross Region Vault
\ "us-vault"
3 / US Cross Region Cold
\ "us-cold"
4 / US Cross Region Flex
\ "us-flex"
5 / US East Region Standard
\ "us-east-standard"
6 / US East Region Vault
\ "us-east-vault"
7 / US East Region Cold
\ "us-east-cold"
8 / US East Region Flex
\ "us-east-flex"
9 / US South Region Standard
\ "us-south-standard"
10 / US South Region Vault
\ "us-south-vault"
[snip]
32 / Toronto Flex
\ "tor01-flex"
location_constraint>1</code></pre>
<olstart="9"type="1">
<li>Specify a canned ACL. IBM Cloud (Storage) supports "public-read" and "private". IBM Cloud(Infra) supports all the canned ACLs. On-Premise COS supports all the canned ACLs.</li>
</ol>
<pre><code>Canned ACL used when creating buckets and/or storing objects in S3.
For more info visit https://docs.aws.amazon.com/AmazonS3/latest/dev/acl-overview.html#canned-acl
Choose a number from below, or type in your own value
1 / Owner gets FULL_CONTROL. No one else has access rights (default). This acl is available on IBM Cloud (Infra), IBM Cloud (Storage), On-Premise COS
\ "private"
2 / Owner gets FULL_CONTROL. The AllUsers group gets READ access. This acl is available on IBM Cloud (Infra), IBM Cloud (Storage), On-Premise IBM COS
\ "public-read"
3 / Owner gets FULL_CONTROL. The AllUsers group gets READ and WRITE access. This acl is available on IBM Cloud (Infra), On-Premise IBM COS
\ "public-read-write"
4 / Owner gets FULL_CONTROL. The AuthenticatedUsers group gets READ access. Not supported on Buckets. This acl is available on IBM Cloud (Infra) and On-Premise IBM COS
\ "authenticated-read"
acl> 1</code></pre>
<olstart="12"type="1">
<li>Review the displayed configuration and accept to save the "remote" then quit. The config file should look like this</li>
<p><ahref="https://www.scaleway.com/object-storage/">Scaleway</a> The Object Storage platform allows you to store anything from backups, logs and web assets to documents and photos. Files can be dropped from the Scaleway console or transferred through our API and CLI or using any S3-compatible tool.</p>
<p>Scaleway provides an S3 interface which can be configured for use with rclone like this:</p>
<p><ahref="https://wasabi.com">Wasabi</a> is a cloud-based object storage service for a broad range of applications and use cases. Wasabi is designed for individuals and organizations that require a high-performance, reliable, and secure data storage infrastructure at minimal cost.</p>
<p>Wasabi provides an S3 interface which can be configured for use with rclone like this.</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
n/s> n
name> wasabi
Type of storage to configure.
Choose a number from below, or type in your own value
[snip]
XX / Amazon S3 (also Dreamhost, Ceph, Minio)
\ "s3"
[snip]
Storage> s3
Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars). Only applies if access_key_id and secret_access_key is blank.
Choose a number from below, or type in your own value
1 / Enter AWS credentials in the next step
\ "false"
2 / Get AWS credentials from the environment (env vars or IAM)
\ "true"
env_auth> 1
AWS Access Key ID - leave blank for anonymous access or runtime credentials.
access_key_id> YOURACCESSKEY
AWS Secret Access Key (password) - leave blank for anonymous access or runtime credentials.
secret_access_key> YOURSECRETACCESSKEY
Region to connect to.
Choose a number from below, or type in your own value
/ The default endpoint - a good choice if you are unsure.
1 | US Region, Northern Virginia, or Pacific Northwest.
| Leave location constraint empty.
\ "us-east-1"
[snip]
region> us-east-1
Endpoint for S3 API.
Leave blank if using AWS to use the default endpoint for the region.
Specify if using an S3 clone such as Ceph.
endpoint> s3.wasabisys.com
Location constraint - must be set to match the Region. Used when creating buckets only.
Choose a number from below, or type in your own value
1 / Empty for US Region, Northern Virginia, or Pacific Northwest.
\ ""
[snip]
location_constraint>
Canned ACL used when creating buckets and/or storing objects in S3.
For more info visit https://docs.aws.amazon.com/AmazonS3/latest/dev/acl-overview.html#canned-acl
Choose a number from below, or type in your own value
1 / Owner gets FULL_CONTROL. No one else has access rights (default).
\ "private"
[snip]
acl>
The server-side encryption algorithm used when storing this object in S3.
Choose a number from below, or type in your own value
1 / None
\ ""
2 / AES256
\ "AES256"
server_side_encryption>
The storage class to use when storing objects in S3.
Choose a number from below, or type in your own value
1 / Default
\ ""
2 / Standard storage class
\ "STANDARD"
3 / Reduced redundancy storage class
\ "REDUCED_REDUNDANCY"
4 / Standard Infrequent Access storage class
\ "STANDARD_IA"
storage_class>
Remote config
--------------------
[wasabi]
env_auth = false
access_key_id = YOURACCESSKEY
secret_access_key = YOURSECRETACCESSKEY
region = us-east-1
endpoint = s3.wasabisys.com
location_constraint =
acl =
server_side_encryption =
storage_class =
--------------------
y) Yes this is OK
e) Edit this remote
d) Delete this remote
y/e/d> y</code></pre>
<p>This will leave the config file looking like this.</p>
<pre><code>[wasabi]
type = s3
provider = Wasabi
env_auth = false
access_key_id = YOURACCESSKEY
secret_access_key = YOURSECRETACCESSKEY
region =
endpoint = s3.wasabisys.com
location_constraint =
acl =
server_side_encryption =
storage_class =</code></pre>
<h3id="alibaba-oss">Alibaba OSS</h3>
<p>Here is an example of making an <ahref="https://www.alibabacloud.com/product/oss/">Alibaba Cloud (Aliyun) OSS</a> configuration. First run:</p>
<pre><code>rclone config</code></pre>
<p>This will guide you through an interactive setup process.</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> oss
Type of storage to configure.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
[snip]
4 / Amazon S3 Compliant Storage Providers including AWS, Alibaba, Ceph, Digital Ocean, Dreamhost, IBM COS, Minio, and Tencent COS
\ "s3"
[snip]
Storage> s3
Choose your S3 provider.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
1 / Amazon Web Services (AWS) S3
\ "AWS"
2 / Alibaba Cloud Object Storage System (OSS) formerly Aliyun
\ "Alibaba"
3 / Ceph Object Storage
\ "Ceph"
[snip]
provider> Alibaba
Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).
Only applies if access_key_id and secret_access_key is blank.
Enter a boolean value (true or false). Press Enter for the default ("false").
Choose a number from below, or type in your own value
1 / Enter AWS credentials in the next step
\ "false"
2 / Get AWS credentials from the environment (env vars or IAM)
\ "true"
env_auth> 1
AWS Access Key ID.
Leave blank for anonymous access or runtime credentials.
Enter a string value. Press Enter for the default ("").
access_key_id> accesskeyid
AWS Secret Access Key (password)
Leave blank for anonymous access or runtime credentials.
Enter a string value. Press Enter for the default ("").
secret_access_key> secretaccesskey
Endpoint for OSS API.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
1 / East China 1 (Hangzhou)
\ "oss-cn-hangzhou.aliyuncs.com"
2 / East China 2 (Shanghai)
\ "oss-cn-shanghai.aliyuncs.com"
3 / North China 1 (Qingdao)
\ "oss-cn-qingdao.aliyuncs.com"
[snip]
endpoint> 1
Canned ACL used when creating buckets and storing or copying objects.
Note that this ACL is applied when server-side copying objects as S3
doesn't copy the ACL from the source but rather writes a fresh one.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
1 / Owner gets FULL_CONTROL. No one else has access rights (default).
\ "private"
2 / Owner gets FULL_CONTROL. The AllUsers group gets READ access.
\ "public-read"
/ Owner gets FULL_CONTROL. The AllUsers group gets READ and WRITE access.
[snip]
acl> 1
The storage class to use when storing new objects in OSS.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
1 / Default
\ ""
2 / Standard storage class
\ "STANDARD"
3 / Archive storage mode.
\ "GLACIER"
4 / Infrequent access storage mode.
\ "STANDARD_IA"
storage_class> 1
Edit advanced config? (y/n)
y) Yes
n) No
y/n> n
Remote config
--------------------
[oss]
type = s3
provider = Alibaba
env_auth = false
access_key_id = accesskeyid
secret_access_key = secretaccesskey
endpoint = oss-cn-hangzhou.aliyuncs.com
acl = private
storage_class = Standard
--------------------
y) Yes this is OK
e) Edit this remote
d) Delete this remote
y/e/d> y</code></pre>
<h3id="tencent-cos">Tencent COS</h3>
<p><ahref="https://intl.cloud.tencent.com/product/cos">Tencent Cloud Object Storage (COS)</a> is a distributed storage service offered by Tencent Cloud for unstructured data. It is secure, stable, massive, convenient, low-delay and low-cost.</p>
<p>To configure access to Tencent COS, follow the steps below:</p>
<oltype="1">
<li>Run <code>rclone config</code> and select <code>n</code> for a new remote.</li>
</ol>
<pre><code>rclone config
No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n</code></pre>
<olstart="2"type="1">
<li>Give the name of the configuration. For example, name it 'cos'.</li>
</ol>
<pre><code>name> cos</code></pre>
<olstart="3"type="1">
<li>Select <code>s3</code> storage.</li>
</ol>
<pre><code>Choose a number from below, or type in your own value
1 / 1Fichier
\ "fichier"
2 / Alias for an existing remote
\ "alias"
3 / Amazon Drive
\ "amazon cloud drive"
4 / Amazon S3 Compliant Storage Providers including AWS, Alibaba, Ceph, Digital Ocean, Dreamhost, IBM COS, Minio, and Tencent COS
\ "s3"
[snip]
Storage> s3</code></pre>
<olstart="4"type="1">
<li>Select <code>TencentCOS</code> provider.</li>
</ol>
<pre><code>Choose a number from below, or type in your own value
1 / Amazon Web Services (AWS) S3
\ "AWS"
[snip]
11 / Tencent Cloud Object Storage (COS)
\ "TencentCOS"
[snip]
provider> TencentCOS</code></pre>
<olstart="5"type="1">
<li>Enter your SecretId and SecretKey of Tencent Cloud.</li>
</ol>
<pre><code>Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).
Only applies if access_key_id and secret_access_key is blank.
Enter a boolean value (true or false). Press Enter for the default ("false").
Choose a number from below, or type in your own value
1 / Enter AWS credentials in the next step
\ "false"
2 / Get AWS credentials from the environment (env vars or IAM)
\ "true"
env_auth> 1
AWS Access Key ID.
Leave blank for anonymous access or runtime credentials.
Enter a string value. Press Enter for the default ("").
access_key_id> AKIDxxxxxxxxxx
AWS Secret Access Key (password)
Leave blank for anonymous access or runtime credentials.
Enter a string value. Press Enter for the default ("").
secret_access_key> xxxxxxxxxxx</code></pre>
<olstart="6"type="1">
<li>Select endpoint for Tencent COS. This is the standard endpoint for different region.</li>
</ol>
<pre><code> 1 / Beijing Region.
\ "cos.ap-beijing.myqcloud.com"
2 / Nanjing Region.
\ "cos.ap-nanjing.myqcloud.com"
3 / Shanghai Region.
\ "cos.ap-shanghai.myqcloud.com"
4 / Guangzhou Region.
\ "cos.ap-guangzhou.myqcloud.com"
[snip]
endpoint> 4</code></pre>
<olstart="7"type="1">
<li>Choose acl and storage class.</li>
</ol>
<pre><code>Note that this ACL is applied when server-side copying objects as S3
doesn't copy the ACL from the source but rather writes a fresh one.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
1 / Owner gets Full_CONTROL. No one else has access rights (default).
\ "default"
[snip]
acl> 1
The storage class to use when storing new objects in Tencent COS.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
<p>For Netease NOS configure as per the configurator <code>rclone config</code> setting the provider <code>Netease</code>. This will automatically set <code>force_path_style = false</code> which is necessary for it to run properly.</p>
<h3id="limitations-4">Limitations</h3>
<p><code>rclone about</code> is not supported by the S3 backend. Backends without this capability cannot determine free space for an rclone mount or use policy <code>mfs</code> (most free space) as a member of an rclone union remote.</p>
<p>See <ahref="https://rclone.org/overview/#optional-features">List of backends that do not support rclone about</a> See <ahref="https://rclone.org/commands/rclone_about/">rclone about</a></p>
<h2id="backblaze-b2">Backblaze B2</h2>
<p>B2 is <ahref="https://www.backblaze.com/b2/">Backblaze's cloud storage system</a>.</p>
<p>Paths are specified as <code>remote:bucket</code> (or <code>remote:</code> for the <code>lsd</code> command.) You may put subdirectories in too, e.g. <code>remote:bucket/path/to/dir</code>.</p>
<p>Here is an example of making a b2 configuration. First run</p>
<pre><code>rclone config</code></pre>
<p>This will guide you through an interactive setup process. To authenticate you will either need your Account ID (a short hex number) and Master Application Key (a long hex number) OR an Application Key, which is the recommended method. See below for further details on generating and using an Application Key.</p>
<pre><code>No remotes found - make a new one
n) New remote
q) Quit config
n/q> n
name> remote
Type of storage to configure.
Choose a number from below, or type in your own value
<p>B2 supports multiple <ahref="https://www.backblaze.com/b2/docs/application_keys.html">Application Keys for different access permission to B2 Buckets</a>.</p>
<p>You can use these with rclone too; you will need to use rclone version 1.43 or later.</p>
<p>Follow Backblaze's docs to create an Application Key with the required permission and add the <code>applicationKeyId</code> as the <code>account</code> and the <code>Application Key</code> itself as the <code>key</code>.</p>
<p>Note that you must put the <em>applicationKeyId</em> as the <code>account</code>– you can't use the master Account ID. If you try then B2 will return 401 errors.</p>
<h3id="fast-list-1">--fast-list</h3>
<p>This remote supports <code>--fast-list</code> which allows you to use fewer transactions in exchange for more memory. See the <ahref="https://rclone.org/docs/#fast-list">rclone docs</a> for more details.</p>
<h3id="modified-time-1">Modified time</h3>
<p>The modified time is stored as metadata on the object as <code>X-Bz-Info-src_last_modified_millis</code> as milliseconds since 1970-01-01 in the Backblaze standard. Other tools should be able to use this as a modified time.</p>
<p>Modified times are used in syncing and are fully supported. Note that if a modification time needs to be updated on an object then it will create a new version of the object.</p>
<p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a> the following characters are also replaced:</p>
<table>
<thead>
<trclass="header">
<th>Character</th>
<thstyle="text-align: center;">Value</th>
<thstyle="text-align: center;">Replacement</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>\</td>
<tdstyle="text-align: center;">0x5C</td>
<tdstyle="text-align: center;">＼</td>
</tr>
</tbody>
</table>
<p>Invalid UTF-8 bytes will also be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as they can't be used in JSON strings.</p>
<p>Note that in 2020-05 Backblaze started allowing characters in file names. Rclone hasn't changed its encoding as this could cause syncs to re-transfer files. If you want rclone not to replace then see the <code>--b2-encoding</code> flag below and remove the <code>BackSlash</code> from the string. This can be set in the config.</p>
<h3id="sha1-checksums">SHA1 checksums</h3>
<p>The SHA1 checksums of the files are checked on upload and download and will be used in the syncing process.</p>
<p>Large files (bigger than the limit in <code>--b2-upload-cutoff</code>) which are uploaded in chunks will store their SHA1 on the object as <code>X-Bz-Info-large_file_sha1</code> as recommended by Backblaze.</p>
<p>For a large file to be uploaded with an SHA1 checksum, the source needs to support SHA1 checksums. The local disk supports SHA1 checksums so large file transfers from local disk will have an SHA1. See <ahref="https://rclone.org/overview/#features">the overview</a> for exactly which remotes support SHA1.</p>
<p>Sources which don't support SHA1, in particular <code>crypt</code> will upload large files without SHA1 checksums. This may be fixed in the future (see <ahref="https://github.com/rclone/rclone/issues/1767">#1767</a>).</p>
<p>Files sizes below <code>--b2-upload-cutoff</code> will always have an SHA1 regardless of the source.</p>
<h3id="transfers">Transfers</h3>
<p>Backblaze recommends that you do lots of transfers simultaneously for maximum speed. In tests from my SSD equipped laptop the optimum setting is about <code>--transfers 32</code> though higher numbers may be used for a slight speed improvement. The optimum number for you may vary depending on your hardware, how big the files are, how much you want to load your computer, etc. The default of <code>--transfers 4</code> is definitely too low for Backblaze B2 though.</p>
<p>Note that uploading big files (bigger than 200 MB by default) will use a 96 MB RAM buffer by default. There can be at most <code>--transfers</code> of these in use at any moment, so this sets the upper limit on the memory used.</p>
<h3id="versions">Versions</h3>
<p>When rclone uploads a new version of a file it creates a <ahref="https://www.backblaze.com/b2/docs/file_versions.html">new version of it</a>. Likewise when you delete a file, the old version will be marked hidden and still be available. Conversely, you may opt in to a "hard delete" of files with the <code>--b2-hard-delete</code> flag which would permanently remove the file instead of hiding it.</p>
<p>Old versions of files, where available, are visible using the <code>--b2-versions</code> flag.</p>
<p><strong>NB</strong> Note that <code>--b2-versions</code> does not work with crypt at the moment <ahref="https://github.com/rclone/rclone/issues/1627">#1627</a>. Using <ahref="https://rclone.org/docs/#backup-dir-dir">--backup-dir</a> with rclone is the recommended way of working around this.</p>
<p>If you wish to remove all the old versions then you can use the <code>rclone cleanup remote:bucket</code> command which will delete all the old versions of files, leaving the current ones intact. You can also supply a path and only old versions under that path will be deleted, e.g. <code>rclone cleanup remote:bucket/path/to/stuff</code>.</p>
<p>Note that <code>cleanup</code> will remove partially uploaded files from the bucket if they are more than a day old.</p>
<p>When you <code>purge</code> a bucket, the current and the old versions will be deleted then the bucket will be deleted.</p>
<p>However <code>delete</code> will cause the current versions of the files to become hidden old versions.</p>
<p>Here is a session showing the listing and retrieval of an old version followed by a <code>cleanup</code> of the old versions.</p>
<p>Show current version and all the versions with <code>--b2-versions</code> flag.</p>
<p>Clean up all the old versions and show that they've gone.</p>
<pre><code>$ rclone -q cleanup b2:cleanup-test
$ rclone -q ls b2:cleanup-test
9 one.txt
$ rclone -q --b2-versions ls b2:cleanup-test
9 one.txt</code></pre>
<h3id="data-usage">Data usage</h3>
<p>It is useful to know how many requests are sent to the server in different scenarios.</p>
<p>All copy commands send the following 4 requests:</p>
<pre><code>/b2api/v1/b2_authorize_account
/b2api/v1/b2_create_bucket
/b2api/v1/b2_list_buckets
/b2api/v1/b2_list_file_names</code></pre>
<p>The <code>b2_list_file_names</code> request will be sent once for every 1k files in the remote path, providing the checksum and modification time of the listed files. As of version 1.33 issue <ahref="https://github.com/rclone/rclone/issues/818">#818</a> causes extra requests to be sent when using B2 with Crypt. When a copy operation does not require any files to be uploaded, no more requests will be sent.</p>
<p>Uploading files that do not require chunking, will send 2 requests per file upload:</p>
<pre><code>/b2api/v1/b2_get_upload_url
/b2api/v1/b2_upload_file/</code></pre>
<p>Uploading files requiring chunking, will send 2 requests (one each to start and finish the upload) and another 2 requests for each chunk:</p>
<pre><code>/b2api/v1/b2_start_large_file
/b2api/v1/b2_get_upload_part_url
/b2api/v1/b2_upload_part/
/b2api/v1/b2_finish_large_file</code></pre>
<h4id="versions-1">Versions</h4>
<p>Versions can be viewed with the <code>--b2-versions</code> flag. When it is set rclone will show and act on older versions of files. For example</p>
<p>Listing without <code>--b2-versions</code></p>
<pre><code>$ rclone -q ls b2:cleanup-test
9 one.txt</code></pre>
<p>And with</p>
<pre><code>$ rclone -q --b2-versions ls b2:cleanup-test
9 one.txt
8 one-v2016-07-04-141032-000.txt
16 one-v2016-07-04-141003-000.txt
15 one-v2016-07-02-155621-000.txt</code></pre>
<p>Showing that the current version is unchanged but older versions can be seen. These have the UTC date that they were uploaded to the server to the nearest millisecond appended to them.</p>
<p>Note that when using <code>--b2-versions</code> no file write operations are permitted, so you can't upload files or delete them.</p>
<h3id="b2-and-rclone-link">B2 and rclone link</h3>
<p>Rclone supports generating file share links for private B2 buckets. They can either be for a file for example:</p>
<pre><code>./rclone link B2:bucket/path/to/file.txt
<p>you can then use the authorization token (the part of the url from the <code>?Authorization=</code> on) on any file path under that directory. For example:</p>
<p>These will be set in the "X-Bz-Test-Mode" header which is documented in the <ahref="https://www.backblaze.com/b2/docs/integration_checklist.html">b2 integrations checklist</a>.</p>
<ul>
<li>Config: test_mode</li>
<li>Env Var: RCLONE_B2_TEST_MODE</li>
<li>Type: string</li>
<li>Default: ""</li>
</ul>
<h4id="b2-versions">--b2-versions</h4>
<p>Include old versions in directory listings. Note that when using this no file write operations are permitted, so you can't upload files or delete them.</p>
<ul>
<li>Config: versions</li>
<li>Env Var: RCLONE_B2_VERSIONS</li>
<li>Type: bool</li>
<li>Default: false</li>
</ul>
<h4id="b2-upload-cutoff">--b2-upload-cutoff</h4>
<p>Cutoff for switching to chunked upload.</p>
<p>Files above this size will be uploaded in chunks of "--b2-chunk-size".</p>
<p>This value should be set no larger than 4.657GiB (== 5GB).</p>
<ul>
<li>Config: upload_cutoff</li>
<li>Env Var: RCLONE_B2_UPLOAD_CUTOFF</li>
<li>Type: SizeSuffix</li>
<li>Default: 200M</li>
</ul>
<h4id="b2-copy-cutoff">--b2-copy-cutoff</h4>
<p>Cutoff for switching to multipart copy</p>
<p>Any files larger than this that need to be server-side copied will be copied in chunks of this size.</p>
<p>The minimum is 0 and the maximum is 4.6GB.</p>
<ul>
<li>Config: copy_cutoff</li>
<li>Env Var: RCLONE_B2_COPY_CUTOFF</li>
<li>Type: SizeSuffix</li>
<li>Default: 4G</li>
</ul>
<h4id="b2-chunk-size">--b2-chunk-size</h4>
<p>Upload chunk size. Must fit in memory.</p>
<p>When uploading large files, chunk the file into this size. Note that these chunks are buffered in memory and there might a maximum of "--transfers" chunks in progress at once. 5,000,000 Bytes is the minimum size.</p>
<p>Disable checksums for large (> upload cutoff) files</p>
<p>Normally rclone will calculate the SHA1 checksum of the input before uploading it so it can add it to metadata on the object. This is great for data integrity checking but can cause long delays for large files to start uploading.</p>
<ul>
<li>Config: disable_checksum</li>
<li>Env Var: RCLONE_B2_DISABLE_CHECKSUM</li>
<li>Type: bool</li>
<li>Default: false</li>
</ul>
<h4id="b2-download-url">--b2-download-url</h4>
<p>Custom endpoint for downloads.</p>
<p>This is usually set to a Cloudflare CDN URL as Backblaze offers free egress for data downloaded through the Cloudflare network. Rclone works with private buckets by sending an "Authorization" header. If the custom endpoint rewrites the requests for authentication, e.g., in Cloudflare Workers, this header needs to be handled properly. Leave blank if you want to use the endpoint provided by Backblaze.</p>
<p>How often internal memory buffer pools will be flushed. Uploads which requires additional buffers (f.e multipart) will use memory pool for allocations. This option controls how often unused buffers will be removed from the pool.</p>
<p><code>rclone about</code> is not supported by the B2 backend. Backends without this capability cannot determine free space for an rclone mount or use policy <code>mfs</code> (most free space) as a member of an rclone union remote.</p>
<p>See <ahref="https://rclone.org/overview/#optional-features">List of backends that do not support rclone about</a> See <ahref="https://rclone.org/commands/rclone_about/">rclone about</a></p>
<h2id="box">Box</h2>
<p>Paths are specified as <code>remote:path</code></p>
<p>Paths may be as deep as required, e.g. <code>remote:directory/subdirectory</code>.</p>
<p>The initial setup for Box involves getting a token from Box which you can do either in your browser, or with a config.json downloaded from Box to use JWT authentication. <code>rclone config</code> walks you through it.</p>
<p>Here is an example of how to make a remote called <code>remote</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> remote
Type of storage to configure.
Choose a number from below, or type in your own value
[snip]
XX / Box
\ "box"
[snip]
Storage> box
Box App Client Id - leave blank normally.
client_id>
Box App Client Secret - leave blank normally.
client_secret>
Box App config.json location
Leave blank normally.
Enter a string value. Press Enter for the default ("").
box_config_file>
Box App Primary Access Token
Leave blank normally.
Enter a string value. Press Enter for the default ("").
<p>See the <ahref="https://rclone.org/remote_setup/">remote setup docs</a> for how to set it up on a machine with no Internet browser available.</p>
<p>Note that rclone runs a webserver on your local machine to collect the token as returned from Box. This only runs from the moment it opens your browser to the moment you get back the verification code. This is on <code>http://127.0.0.1:53682/</code> and this it may require you to unblock it temporarily if you are running a host firewall.</p>
<p>Once configured you can then use <code>rclone</code> like this,</p>
<p>List directories in top level of your Box</p>
<pre><code>rclone lsd remote:</code></pre>
<p>List all the files in your Box</p>
<pre><code>rclone ls remote:</code></pre>
<p>To copy a local directory to an Box directory called backup</p>
<h3id="using-rclone-with-an-enterprise-account-with-sso">Using rclone with an Enterprise account with SSO</h3>
<p>If you have an "Enterprise" account type with Box with single sign on (SSO), you need to create a password to use Box with rclone. This can be done at your Enterprise Box account by going to Settings, "Account" Tab, and then set the password in the "Authentication" field.</p>
<p>Once you have done this, you can setup your Enterprise Box account using the same procedure detailed above in the, using the password you have just set.</p>
<p>According to the <ahref="https://developer.box.com/v2.0/docs/oauth-20#section-6-using-the-access-and-refresh-tokens">box docs</a>:</p>
<blockquote>
<p>Each refresh_token is valid for one use in 60 days.</p>
</blockquote>
<p>This means that if you</p>
<ul>
<li>Don't use the box remote for 60 days</li>
<li>Copy the config file with a box refresh token in and use it in two places</li>
<li>Get an error on a token refresh</li>
</ul>
<p>then rclone will return an error which includes the text <code>Invalid refresh token</code>.</p>
<p>To fix this you will need to use oauth2 again to update the refresh token. You can use the methods in <ahref="https://rclone.org/remote_setup/">the remote setup docs</a>, bearing in mind that if you use the copy the config file method, you should not use that remote on the computer you did the authentication on.</p>
<p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a> the following characters are also replaced:</p>
<table>
<thead>
<trclass="header">
<th>Character</th>
<thstyle="text-align: center;">Value</th>
<thstyle="text-align: center;">Replacement</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>\</td>
<tdstyle="text-align: center;">0x5C</td>
<tdstyle="text-align: center;">＼</td>
</tr>
</tbody>
</table>
<p>File names can also not end with the following characters. These only get replaced if they are the last character in the name:</p>
<table>
<thead>
<trclass="header">
<th>Character</th>
<thstyle="text-align: center;">Value</th>
<thstyle="text-align: center;">Replacement</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>SP</td>
<tdstyle="text-align: center;">0x20</td>
<tdstyle="text-align: center;">␠</td>
</tr>
</tbody>
</table>
<p>Invalid UTF-8 bytes will also be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as they can't be used in JSON strings.</p>
<h3id="transfers-1">Transfers</h3>
<p>For files above 50MB rclone will use a chunked transfer. Rclone will upload up to <code>--transfers</code> chunks at the same time (shared among all the multipart uploads). Chunks are buffered in memory and are normally 8MB so increasing <code>--transfers</code> will increase memory use.</p>
<h3id="deleting-files-1">Deleting files</h3>
<p>Depending on the enterprise settings for your user, the item will either be actually deleted from Box or moved to the trash.</p>
<p>Emptying the trash is supported via the rclone however cleanup command however this deletes every trashed file and folder individually so it may take a very long time. Emptying the trash via the WebUI does not have this limitation so it is advised to empty the trash via the WebUI.</p>
<h3id="root-folder-id">Root folder ID</h3>
<p>You can set the <code>root_folder_id</code> for rclone. This is the directory (identified by its <code>Folder ID</code>) that rclone considers to be the root of your Box drive.</p>
<p>Normally you will leave this blank and rclone will determine the correct root to use itself.</p>
<p>However you can set this to restrict rclone to a specific folder hierarchy.</p>
<p>In order to do this you will have to find the <code>Folder ID</code> of the directory you wish rclone to display. This will be the last segment of the URL when you open the relevant folder in the Box web interface.</p>
<p>So if the folder you want rclone to use has a URL which looks like <code>https://app.box.com/folder/11xxxxxxxxx8</code> in the browser, then you use <code>11xxxxxxxxx8</code> as the <code>root_folder_id</code> in the config.</p>
<h3id="standard-options-5">Standard Options</h3>
<p>Here are the standard options specific to box (Box).</p>
<p>Note that Box is case insensitive so you can't have a file called "Hello.doc" and one called "hello.doc".</p>
<p>Box file names can't have the <code>\</code> character in. rclone maps this to and from an identical looking unicode equivalent <code>＼</code> (U+FF3C Fullwidth Reverse Solidus).</p>
<p>Box only supports filenames up to 255 characters in length.</p>
<p><code>rclone about</code> is not supported by the Box backend. Backends without this capability cannot determine free space for an rclone mount or use policy <code>mfs</code> (most free space) as a member of an rclone union remote.</p>
<p>See <ahref="https://rclone.org/overview/#optional-features">List of backends that do not support rclone about</a> See <ahref="https://rclone.org/commands/rclone_about/">rclone about</a></p>
<h2id="cache-beta">Cache (BETA)</h2>
<p>The <code>cache</code> remote wraps another existing remote and stores file structure and its data for long running tasks like <code>rclone mount</code>.</p>
<h2id="status-1">Status</h2>
<p>The cache backend code is working but it currently doesn't have a maintainer so there are <ahref="https://github.com/rclone/rclone/issues?q=is%3Aopen+is%3Aissue+label%3Abug+label%3A%22Remote%3A+Cache%22">outstanding bugs</a> which aren't getting fixed.</p>
<p>The cache backend is due to be phased out in favour of the VFS caching layer eventually which is more tightly integrated into rclone.</p>
<p>Until this happens we recommend only using the cache backend if you find you can't work without it. There are many docs online describing the use of the cache backend to minimize API hits and by-and-large these are out of date and the cache backend isn't needed in those scenarios any more.</p>
<h2id="setup-1">Setup</h2>
<p>To get started you just need to have an existing remote which can be configured with <code>cache</code>.</p>
<p>Here is an example of how to make a remote called <code>test-cache</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>No remotes found - make a new one
n) New remote
r) Rename remote
c) Copy remote
s) Set configuration password
q) Quit config
n/r/c/s/q> n
name> test-cache
Type of storage to configure.
Choose a number from below, or type in your own value
[snip]
XX / Cache a remote
\ "cache"
[snip]
Storage> cache
Remote to cache.
Normally should contain a ':' and a path, e.g. "myremote:path/to/dir",
"myremote:bucket" or maybe "myremote:" (not recommended).
remote> local:/test
Optional: The URL of the Plex server
plex_url> http://127.0.0.1:32400
Optional: The username of the Plex user
plex_username> dummyusername
Optional: The password of the Plex user
y) Yes type in my own password
g) Generate random password
n) No leave this optional password blank
y/g/n> y
Enter the password:
password:
Confirm the password:
password:
The size of a chunk. Lower value good for slow connections but can affect seamless reading.
Default: 5M
Choose a number from below, or type in your own value
1 / 1MB
\ "1m"
2 / 5 MB
\ "5M"
3 / 10 MB
\ "10M"
chunk_size> 2
How much time should object info (file size, file hashes, etc.) be stored in cache. Use a very high value if you don't plan on changing the source FS from outside the cache.
Accepted units are: "s", "m", "h".
Default: 5m
Choose a number from below, or type in your own value
1 / 1 hour
\ "1h"
2 / 24 hours
\ "24h"
3 / 24 hours
\ "48h"
info_age> 2
The maximum size of stored chunks. When the storage grows beyond this size, the oldest chunks will be deleted.
Default: 10G
Choose a number from below, or type in your own value
1 / 500 MB
\ "500M"
2 / 1 GB
\ "1G"
3 / 10 GB
\ "10G"
chunk_total_size> 3
Remote config
--------------------
[test-cache]
remote = local:/test
plex_url = http://127.0.0.1:32400
plex_username = dummyusername
plex_password = *** ENCRYPTED ***
chunk_size = 5M
info_age = 48h
chunk_total_size = 10G</code></pre>
<p>You can then use it like this,</p>
<p>List directories in top level of your drive</p>
<pre><code>rclone lsd test-cache:</code></pre>
<p>List all the files in your drive</p>
<pre><code>rclone ls test-cache:</code></pre>
<p>To start a cached mount</p>
<pre><code>rclone mount --allow-other test-cache: /var/tmp/test-cache</code></pre>
<h3id="write-features">Write Features</h3>
<h3id="offline-uploading">Offline uploading</h3>
<p>In an effort to make writing through cache more reliable, the backend now supports this feature which can be activated by specifying a <code>cache-tmp-upload-path</code>.</p>
<p>A files goes through these states when using this feature:</p>
<oltype="1">
<li>An upload is started (usually by copying a file on the cache remote)</li>
<li>When the copy to the temporary location is complete the file is part of the cached remote and looks and behaves like any other file (reading included)</li>
<li>After <code>cache-tmp-wait-time</code> passes and the file is next in line, <code>rclone move</code> is used to move the file to the cloud provider</li>
<li>Reading the file still works during the upload but most modifications on it will be prohibited</li>
<li>Once the move is complete the file is unlocked for modifications as it becomes as any other regular file</li>
<li>If the file is being read through <code>cache</code> when it's actually deleted from the temporary path then <code>cache</code> will simply swap the source to the cloud provider without interrupting the reading (small blip can happen though)</li>
<p>Files are uploaded in sequence and only one file is uploaded at a time. Uploads will be stored in a queue and be processed based on the order they were added. The queue and the temporary storage is persistent across restarts but can be cleared on startup with the <code>--cache-db-purge</code> flag.</p>
<h3id="write-support">Write Support</h3>
<p>Writes are supported through <code>cache</code>. One caveat is that a mounted cache remote does not add any retry or fallback mechanism to the upload operation. This will depend on the implementation of the wrapped remote. Consider using <code>Offline uploading</code> for reliable writes.</p>
<p>One special case is covered with <code>cache-writes</code> which will cache the file data at the same time as the upload when it is enabled making it available from the cache store immediately once the upload is finished.</p>
<p>To counter the high latency between a local PC where rclone is running and cloud providers, the cache remote can split multiple requests to the cloud provider for smaller file chunks and combines them together locally where they can be available almost immediately before the reader usually needs them.</p>
<p>This is similar to buffering when media files are played online. Rclone will stay around the current marker but always try its best to stay ahead and prepare the data before.</p>
<h4id="plex-integration">Plex Integration</h4>
<p>There is a direct integration with Plex which allows cache to detect during reading if the file is in playback or not. This helps cache to adapt how it queries the cloud provider depending on what is needed for.</p>
<p>Scans will have a minimum amount of workers (1) while in a confirmed playback cache will deploy the configured number of workers.</p>
<p>This integration opens the doorway to additional performance improvements which will be explored in the near future.</p>
<p><strong>Note:</strong> If Plex options are not configured, <code>cache</code> will function with its configured options without adapting any of its settings.</p>
<p>How to enable? Run <code>rclone config</code> and add all the Plex options (endpoint, username and password) in your remote and it will be automatically enabled.</p>
<p>Affected settings: - <code>cache-workers</code>: <em>Configured value</em> during confirmed playback or <em>1</em> all the other times</p>
<p>When the Plex server is configured to only accept secure connections, it is possible to use <code>.plex.direct</code> URLs to ensure certificate validation succeeds. These URLs are used by Plex internally to connect to the Plex server securely.</p>
<p>The format for these URLs is the following:</p>
<p>The <code>ip-with-dots-replaced</code> part can be any IPv4 address, where the dots have been replaced with dashes, e.g. <code>127.0.0.1</code> becomes <code>127-0-0-1</code>.</p>
<p>To get the <code>server-hash</code> part, the easiest way is to visit</p>
<p>This page will list all the available Plex servers for your account with at least one <code>.plex.direct</code> link for each. Copy one URL and replace the IP address with the desired address. This can be used as the <code>plex_url</code> value.</p>
<h3id="known-issues">Known issues</h3>
<h4id="mount-and---dir-cache-time">Mount and --dir-cache-time</h4>
<p>--dir-cache-time controls the first layer of directory caching which works at the mount layer. Being an independent caching mechanism from the <code>cache</code> backend, it will manage its own entries based on the configured time.</p>
<p>To avoid getting in a scenario where dir cache has obsolete data and cache would have the correct one, try to set <code>--dir-cache-time</code> to a lower time than <code>--cache-info-age</code>. Default values are already configured in this way.</p>
<h4id="windows-support---experimental">Windows support - Experimental</h4>
<p>There are a couple of issues with Windows <code>mount</code> functionality that still require some investigations. It should be considered as experimental thus far as fixes come in for this OS.</p>
<p>Most of the issues seem to be related to the difference between filesystems on Linux flavors and Windows as cache is heavily dependent on them.</p>
<p>Any reports or feedback on how cache behaves on this OS is greatly appreciated.</p>
<h4id="risk-of-throttling">Risk of throttling</h4>
<p>Future iterations of the cache backend will make use of the pooling functionality of the cloud provider to synchronize and at the same time make writing through it more tolerant to failures.</p>
<p>There are a couple of enhancements in track to add these but in the meantime there is a valid concern that the expiring cache listings can lead to cloud provider throttles or bans due to repeated queries on it for very large mounts.</p>
<p>Some recommendations: - don't use a very small interval for entry information (<code>--cache-info-age</code>) - while writes aren't yet optimised, you can still write through <code>cache</code> which gives you the advantage of adding the file in the cache at the same time if configured to do so.</p>
<p>One common scenario is to keep your data encrypted in the cloud provider using the <code>crypt</code> remote. <code>crypt</code> uses a similar technique to wrap around an existing remote and handles this translation in a seamless way.</p>
<p>There is an issue with wrapping the remotes in this order: <strong>cloud remote</strong> -><strong>crypt</strong> -><strong>cache</strong></p>
<p>During testing, I experienced a lot of bans with the remotes in this order. I suspect it might be related to how crypt opens files on the cloud provider which makes it think we're downloading the full file instead of small chunks. Organizing the remotes in this order yields better results: <strong>cloud remote</strong> -><strong>cache</strong> -><strong>crypt</strong></p>
<p><code>cache</code> can not differentiate between relative and absolute paths for the wrapped remote. Any path given in the <code>remote</code> config setting and on the command line will be passed to the wrapped remote as is, but for storing the chunks on disk the path will be made relative by removing any leading <code>/</code> character.</p>
<p>This behavior is irrelevant for most backend types, but there are backends where a leading <code>/</code> changes the effective directory, e.g. in the <code>sftp</code> backend paths starting with a <code>/</code> are relative to the root of the SSH server and paths without are relative to the user home directory. As a result <code>sftp:bin</code> and <code>sftp:/bin</code> will share the same cache folder, even if they represent a different directory on the SSH server.</p>
<h3id="cache-and-remote-control---rc">Cache and Remote Control (--rc)</h3>
<p>Cache supports the new <code>--rc</code> mode in rclone and can be remote controlled through the following end points: By default, the listener is disabled if you do not add the flag.</p>
<h3id="rc-cacheexpire">rc cache/expire</h3>
<p>Purge a remote from the cache backend. Supports either a directory or a file. It supports both encrypted and unencrypted file names if cache is wrapped by crypt.</p>
<p>Params: - <strong>remote</strong> = path to remote <strong>(required)</strong> - <strong>withData</strong> = true/false to delete cached data (chunks) as well <em>(optional, false by default)</em></p>
<h3id="standard-options-6">Standard Options</h3>
<p>Here are the standard options specific to cache (Cache a remote).</p>
<h4id="cache-remote">--cache-remote</h4>
<p>Remote to cache. Normally should contain a ':' and a path, e.g. "myremote:path/to/dir", "myremote:bucket" or maybe "myremote:" (not recommended).</p>
<p><strong>NB</strong> Input to this must be obscured - see <ahref="https://rclone.org/commands/rclone_obscure/">rclone obscure</a>.</p>
<ul>
<li>Config: plex_password</li>
<li>Env Var: RCLONE_CACHE_PLEX_PASSWORD</li>
<li>Type: string</li>
<li>Default: ""</li>
</ul>
<h4id="cache-chunk-size">--cache-chunk-size</h4>
<p>The size of a chunk (partial file data).</p>
<p>Use lower numbers for slower connections. If the chunk size is changed, any downloaded chunks will be invalid and cache-chunk-path will need to be cleared or unexpected EOF errors will occur.</p>
<ul>
<li>Config: chunk_size</li>
<li>Env Var: RCLONE_CACHE_CHUNK_SIZE</li>
<li>Type: SizeSuffix</li>
<li>Default: 5M</li>
<li>Examples:
<ul>
<li>"1m"
<ul>
<li>1MB</li>
</ul></li>
<li>"5M"
<ul>
<li>5 MB</li>
</ul></li>
<li>"10M"
<ul>
<li>10 MB</li>
</ul></li>
</ul></li>
</ul>
<h4id="cache-info-age">--cache-info-age</h4>
<p>How long to cache file structure information (directory listings, file size, times, etc.). If all write operations are done through the cache then you can safely make this value very large as the cache store will also be updated in real time.</p>
<p>Path to where partial file data (chunks) are stored locally. The remote name is appended to the final path.</p>
<p>This config follows the "--cache-db-path". If you specify a custom location for "--cache-db-path" and don't specify one for "--cache-chunk-path" then "--cache-chunk-path" will use the same path as "--cache-db-path".</p>
<p>How often should the cache perform cleanups of the chunk storage. The default value should be ok for most people. If you find that the cache goes over "cache-chunk-total-size" too often then try to lower this value to force it to perform cleanups more often.</p>
<p>How many times to retry a read from a cache storage.</p>
<p>Since reading from a cache stream is independent from downloading file data, readers can get to a point where there's no more data in the cache. Most of the times this can indicate a connectivity issue if cache isn't able to provide file data anymore.</p>
<p>For really slow connections, increase this to a point where the stream is able to provide data but your experience will be very stuttering.</p>
<ul>
<li>Config: read_retries</li>
<li>Env Var: RCLONE_CACHE_READ_RETRIES</li>
<li>Type: int</li>
<li>Default: 10</li>
</ul>
<h4id="cache-workers">--cache-workers</h4>
<p>How many workers should run in parallel to download chunks.</p>
<p>Higher values will mean more parallel processing (better CPU needed) and more concurrent requests on the cloud provider. This impacts several aspects like the cloud provider API limits, more stress on the hardware that rclone runs on but it also means that streams will be more fluid and data will be available much more faster to readers.</p>
<p><strong>Note</strong>: If the optional Plex integration is enabled then this setting will adapt to the type of reading performed and the value specified here will be used as a maximum number of workers to use.</p>
<p>Disable the in-memory cache for storing chunks during streaming.</p>
<p>By default, cache will keep file data during streaming in RAM as well to provide it to readers as fast as possible.</p>
<p>This transient data is evicted as soon as it is read and the number of chunks stored doesn't exceed the number of workers. However, depending on other settings like "cache-chunk-size" and "cache-workers" this footprint can increase if there are parallel streams too (multiple files being read at the same time).</p>
<p>If the hardware permits it, use this feature to provide an overall better performance during streaming but it can also be disabled if RAM is not available on the local machine.</p>
<ul>
<li>Config: chunk_no_memory</li>
<li>Env Var: RCLONE_CACHE_CHUNK_NO_MEMORY</li>
<li>Type: bool</li>
<li>Default: false</li>
</ul>
<h4id="cache-rps">--cache-rps</h4>
<p>Limits the number of requests per second to the source FS (-1 to disable)</p>
<p>This setting places a hard limit on the number of requests per second that cache will be doing to the cloud provider remote and try to respect that value by setting waits between reads.</p>
<p>If you find that you're getting banned or limited on the cloud provider through cache and know that a smaller number of requests per second will allow you to work with it then you can use this setting for that.</p>
<p>A good balance of all the other settings should make this setting useless but it is available to set for more special cases.</p>
<p><strong>NOTE</strong>: This will limit the number of requests during streams but other API calls to the cloud provider like directory listings will still pass.</p>
<ul>
<li>Config: rps</li>
<li>Env Var: RCLONE_CACHE_RPS</li>
<li>Type: int</li>
<li>Default: -1</li>
</ul>
<h4id="cache-writes">--cache-writes</h4>
<p>Cache file data on writes through the FS</p>
<p>If you need to read files immediately after you upload them through cache you can enable this flag to have their data stored in the cache store at the same time during upload.</p>
<p>How long to wait for the DB to be available - 0 is unlimited</p>
<p>Only one process can have the DB open at any one time, so rclone waits for this duration for the DB to become available before it gives an error.</p>
<p>If you set it to 0 then it will wait forever.</p>
<ul>
<li>Config: db_wait_time</li>
<li>Env Var: RCLONE_CACHE_DB_WAIT_TIME</li>
<li>Type: Duration</li>
<li>Default: 1s</li>
</ul>
<h3id="backend-commands-1">Backend commands</h3>
<p>Here are the commands specific to the cache backend.</p>
<p>The <code>chunker</code> overlay transparently splits large files into smaller chunks during upload to wrapped remote and transparently assembles them back when the file is downloaded. This allows to effectively overcome size limits imposed by storage providers.</p>
<p>To use it, first set up the underlying remote following the configuration instructions for that remote. You can also use a local pathname instead of a remote.</p>
<p>First check your chosen remote is working - we'll call it <code>remote:path</code> here. Note that anything inside <code>remote:path</code> will be chunked and anything outside won't. This means that if you are using a bucket based remote (e.g. S3, B2, swift) then you should probably put the bucket in the remote <code>s3:bucket</code>.</p>
<p>Now configure <code>chunker</code> using <code>rclone config</code>. We will call this one <code>overlay</code> to separate it from the <code>remote</code> itself.</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> overlay
Type of storage to configure.
Choose a number from below, or type in your own value
[snip]
XX / Transparently chunk/split large files
\ "chunker"
[snip]
Storage> chunker
Remote to chunk/unchunk.
Normally should contain a ':' and a path, e.g. "myremote:path/to/dir",
"myremote:bucket" or maybe "myremote:" (not recommended).
Enter a string value. Press Enter for the default ("").
remote> remote:path
Files larger than chunk size will be split in chunks.
Enter a size with suffix k,M,G,T. Press Enter for the default ("2G").
chunk_size> 100M
Choose how chunker handles hash sums. All modes but "none" require metadata.
Enter a string value. Press Enter for the default ("md5").
Choose a number from below, or type in your own value
1 / Pass any hash supported by wrapped remote for non-chunked files, return nothing otherwise
\ "none"
2 / MD5 for composite files
\ "md5"
3 / SHA1 for composite files
\ "sha1"
4 / MD5 for all files
\ "md5all"
5 / SHA1 for all files
\ "sha1all"
6 / Copying a file to chunker will request MD5 from the source falling back to SHA1 if unsupported
\ "md5quick"
7 / Similar to "md5quick" but prefers SHA1 over MD5
\ "sha1quick"
hash_type> md5
Edit advanced config? (y/n)
y) Yes
n) No
y/n> n
Remote config
--------------------
[overlay]
type = chunker
remote = remote:bucket
chunk_size = 100M
hash_type = md5
--------------------
y) Yes this is OK
e) Edit this remote
d) Delete this remote
y/e/d> y</code></pre>
<h3id="specifying-the-remote">Specifying the remote</h3>
<p>In normal use, make sure the remote has a <code>:</code> in. If you specify the remote without a <code>:</code> then rclone will use a local directory of that name. So if you use a remote of <code>/path/to/secret/files</code> then rclone will chunk stuff in that directory. If you use a remote of <code>name</code> then rclone will put files in a directory called <code>name</code> in the current directory.</p>
<h3id="chunking">Chunking</h3>
<p>When rclone starts a file upload, chunker checks the file size. If it doesn't exceed the configured chunk size, chunker will just pass the file to the wrapped remote. If a file is large, chunker will transparently cut data in pieces with temporary names and stream them one by one, on the fly. Each data chunk will contain the specified number of bytes, except for the last one which may have less data. If file size is unknown in advance (this is called a streaming upload), chunker will internally create a temporary copy, record its size and repeat the above process.</p>
<p>When upload completes, temporary chunk files are finally renamed. This scheme guarantees that operations can be run in parallel and look from outside as atomic. A similar method with hidden temporary chunks is used for other operations (copy/move/rename, etc.). If an operation fails, hidden chunks are normally destroyed, and the target composite file stays intact.</p>
<p>When a composite file download is requested, chunker transparently assembles it by concatenating data chunks in order. As the split is trivial one could even manually concatenate data chunks together to obtain the original content.</p>
<p>When the <code>list</code> rclone command scans a directory on wrapped remote, the potential chunk files are accounted for, grouped and assembled into composite directory entries. Any temporary chunks are hidden.</p>
<p>List and other commands can sometimes come across composite files with missing or invalid chunks, e.g. shadowed by like-named directory or another file. This usually means that wrapped file system has been directly tampered with or damaged. If chunker detects a missing chunk it will by default print warning, skip the whole incomplete group of chunks but proceed with current command. You can set the <code>--chunker-fail-hard</code> flag to have commands abort with error message in such cases.</p>
<h4id="chunk-names">Chunk names</h4>
<p>The default chunk name format is <code>*.rclone_chunk.###</code>, hence by default chunk names are <code>BIG_FILE_NAME.rclone_chunk.001</code>, <code>BIG_FILE_NAME.rclone_chunk.002</code> etc. You can configure another name format using the <code>name_format</code> configuration file option. The format uses asterisk <code>*</code> as a placeholder for the base file name and one or more consecutive hash characters <code>#</code> as a placeholder for sequential chunk number. There must be one and only one asterisk. The number of consecutive hash characters defines the minimum length of a string representing a chunk number. If decimal chunk number has less digits than the number of hashes, it is left-padded by zeros. If the decimal string is longer, it is left intact. By default numbering starts from 1 but there is another option that allows user to start from 0, e.g. for compatibility with legacy software.</p>
<p>For example, if name format is <code>big_*-##.part</code> and original file name is <code>data.txt</code> and numbering starts from 0, then the first chunk will be named <code>big_data.txt-00.part</code>, the 99th chunk will be <code>big_data.txt-98.part</code> and the 302nd chunk will become <code>big_data.txt-301.part</code>.</p>
<p>Note that <code>list</code> assembles composite directory entries only when chunk names match the configured format and treats non-conforming file names as normal non-chunked files.</p>
<p>When using <code>norename</code> transactions, chunk names will additionally have a unique file version suffix. For example, <code>BIG_FILE_NAME.rclone_chunk.001_bp562k</code>.</p>
<p>Besides data chunks chunker will by default create metadata object for a composite file. The object is named after the original file. Chunker allows user to disable metadata completely (the <code>none</code> format). Note that metadata is normally not created for files smaller than the configured chunk size. This may change in future rclone releases.</p>
<p>There is no field for composite file name as it's simply equal to the name of meta object on the wrapped remote. Please refer to respective sections for details on hashsums and modified time handling.</p>
<h4id="no-metadata">No metadata</h4>
<p>You can disable meta objects by setting the meta format option to <code>none</code>. In this mode chunker will scan directory for all files that follow configured chunk name format, group them by detecting chunks with the same base name and show group names as virtual composite files. This method is more prone to missing chunk errors (especially missing last chunk) than format with metadata enabled.</p>
<h3id="hashsums">Hashsums</h3>
<p>Chunker supports hashsums only when a compatible metadata is present. Hence, if you choose metadata format of <code>none</code>, chunker will report hashsum as <code>UNSUPPORTED</code>.</p>
<p>Please note that by default metadata is stored only for composite files. If a file is smaller than configured chunk size, chunker will transparently redirect hash requests to wrapped remote, so support depends on that. You will see the empty string as a hashsum of requested type for small files if the wrapped remote doesn't support it.</p>
<p>Many storage backends support MD5 and SHA1 hash types, so does chunker. With chunker you can choose one or another but not both. MD5 is set by default as the most supported type. Since chunker keeps hashes for composite files and falls back to the wrapped remote hash for non-chunked ones, we advise you to choose the same hash type as supported by wrapped remote so that your file listings look coherent.</p>
<p>If your storage backend does not support MD5 or SHA1 but you need consistent file hashing, configure chunker with <code>md5all</code> or <code>sha1all</code>. These two modes guarantee given hash for all files. If wrapped remote doesn't support it, chunker will then add metadata to all files, even small. However, this can double the amount of small files in storage and incur additional service charges. You can even use chunker to force md5/sha1 support in any other remote at expense of sidecar meta objects by setting e.g. <code>chunk_type=sha1all</code> to force hashsums and <code>chunk_size=1P</code> to effectively disable chunking.</p>
<p>Normally, when a file is copied to chunker controlled remote, chunker will ask the file source for compatible file hash and revert to on-the-fly calculation if none is found. This involves some CPU overhead but provides a guarantee that given hashsum is available. Also, chunker will reject a server-side copy or move operation if source and destination hashsum types are different resulting in the extra network bandwidth, too. In some rare cases this may be undesired, so chunker provides two optional choices: <code>sha1quick</code> and <code>md5quick</code>. If the source does not support primary hash type and the quick mode is enabled, chunker will try to fall back to the secondary type. This will save CPU and bandwidth but can result in empty hashsums at destination. Beware of consequences: the <code>sync</code> command will revert (sometimes silently) to time/size comparison if compatible hashsums between source and target are not found.</p>
<h3id="modified-time-2">Modified time</h3>
<p>Chunker stores modification times using the wrapped remote so support depends on that. For a small non-chunked file the chunker overlay simply manipulates modification time of the wrapped remote file. For a composite file with metadata chunker will get and set modification time of the metadata object on the wrapped remote. If file is chunked but metadata format is <code>none</code> then chunker will use modification time of the first data chunk.</p>
<li>Collect all your chunked files under a directory and have your chunker remote point to it.</li>
<li>Create another directory (most probably on the same cloud storage) and configure a new remote with desired metadata format, hash type, chunk naming etc.</li>
<li>Now run <code>rclone sync -i oldchunks: newchunks:</code> and all your data will be transparently converted in transfer. This may take some time, yet chunker will try server-side copy if possible.</li>
<li>After checking data integrity you may remove configuration section of the old remote.</li>
</ul>
<p>If rclone gets killed during a long operation on a big composite file, hidden temporary chunks may stay in the directory. They will not be shown by the <code>list</code> command but will eat up your account quota. Please note that the <code>deletefile</code> command deletes only active chunks of a file. As a workaround, you can use remote of the wrapped file system to see them. An easy way to get rid of hidden garbage is to copy littered directory somewhere using the chunker remote and purge the original directory. The <code>copy</code> command will copy only active chunks while the <code>purge</code> will remove everything including garbage.</p>
<h3id="caveats-and-limitations">Caveats and Limitations</h3>
<p>Chunker requires wrapped remote to support server-side <code>move</code> (or <code>copy</code> + <code>delete</code>) operations, otherwise it will explicitly refuse to start. This is because it internally renames temporary chunk files to their final names when an operation completes successfully.</p>
<p>Chunker encodes chunk number in file name, so with default <code>name_format</code> setting it adds 17 characters. Also chunker adds 7 characters of temporary suffix during operations. Many file systems limit base file name without path by 255 characters. Using rclone's crypt remote as a base file system limits file name by 143 characters. Thus, maximum name length is 231 for most files and 119 for chunker-over-crypt. A user in need can change name format to e.g. <code>*.rcc##</code> and save 10 characters (provided at most 99 chunks per file).</p>
<p>Note that a move implemented using the copy-and-delete method may incur double charging with some cloud storage providers.</p>
<p>Chunker will not automatically rename existing chunks when you run <code>rclone config</code> on a live remote and change the chunk name format. Beware that in result of this some files which have been treated as chunks before the change can pop up in directory listings as normal files and vice versa. The same warning holds for the chunk size. If you desperately need to change critical chunking settings, you should run data migration as described above.</p>
<p>If wrapped remote is case insensitive, the chunker overlay will inherit that property (so you can't have a file called "Hello.doc" and "hello.doc" in the same directory).</p>
<p>Chunker included in rclone releases up to <code>v1.54</code> can sometimes fail to detect metadata produced by recent versions of rclone. We recommend users to keep rclone up-to-date to avoid data corruption.</p>
<p>Here are the standard options specific to chunker (Transparently chunk/split large files).</p>
<h4id="chunker-remote">--chunker-remote</h4>
<p>Remote to chunk/unchunk. Normally should contain a ':' and a path, e.g. "myremote:path/to/dir", "myremote:bucket" or maybe "myremote:" (not recommended).</p>
<p>String format of chunk file names. The two placeholders are: base file name (*) and chunk number (#...). There must be one and only one asterisk and one or more consecutive hash characters. If chunk number has less digits than the number of hashes, it is left-padded by zeros. If there are more digits in the number, they are left as is. Possible chunk files are ignored if their name does not match given format.</p>
<p><ahref="https://sharefile.com">Citrix ShareFile</a> is a secure file sharing and transfer service aimed as business.</p>
<p>The initial setup for Citrix ShareFile involves getting a token from Citrix ShareFile which you can in your browser. <code>rclone config</code> walks you through it.</p>
<p>Here is an example of how to make a remote called <code>remote</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> remote
Type of storage to configure.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
XX / Citrix Sharefile
\ "sharefile"
Storage> sharefile
** See help for sharefile backend at: https://rclone.org/sharefile/ **
<p>See the <ahref="https://rclone.org/remote_setup/">remote setup docs</a> for how to set it up on a machine with no Internet browser available.</p>
<p>Note that rclone runs a webserver on your local machine to collect the token as returned from Citrix ShareFile. This only runs from the moment it opens your browser to the moment you get back the verification code. This is on <code>http://127.0.0.1:53682/</code> and this it may require you to unblock it temporarily if you are running a host firewall.</p>
<p>Once configured you can then use <code>rclone</code> like this,</p>
<p>List directories in top level of your ShareFile</p>
<pre><code>rclone lsd remote:</code></pre>
<p>List all the files in your ShareFile</p>
<pre><code>rclone ls remote:</code></pre>
<p>To copy a local directory to an ShareFile directory called backup</p>
<p>Paths may be as deep as required, e.g. <code>remote:directory/subdirectory</code>.</p>
<h3id="modified-time-and-hashes-2">Modified time and hashes</h3>
<p>ShareFile allows modification times to be set on objects accurate to 1 second. These will be used to detect whether objects need syncing or not.</p>
<p>ShareFile supports MD5 type hashes, so you can use the <code>--checksum</code> flag.</p>
<h3id="transfers-2">Transfers</h3>
<p>For files above 128MB rclone will use a chunked transfer. Rclone will upload up to <code>--transfers</code> chunks at the same time (shared among all the multipart uploads). Chunks are buffered in memory and are normally 64MB so increasing <code>--transfers</code> will increase memory use.</p>
<h3id="limitations-7">Limitations</h3>
<p>Note that ShareFile is case insensitive so you can't have a file called "Hello.doc" and one called "hello.doc".</p>
<p>ShareFile only supports filenames up to 256 characters in length.</p>
<p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a> the following characters are also replaced:</p>
<table>
<thead>
<trclass="header">
<th>Character</th>
<thstyle="text-align: center;">Value</th>
<thstyle="text-align: center;">Replacement</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>\</td>
<tdstyle="text-align: center;">0x5C</td>
<tdstyle="text-align: center;">＼</td>
</tr>
<trclass="even">
<td>*</td>
<tdstyle="text-align: center;">0x2A</td>
<tdstyle="text-align: center;">＊</td>
</tr>
<trclass="odd">
<td><</td>
<tdstyle="text-align: center;">0x3C</td>
<tdstyle="text-align: center;">＜</td>
</tr>
<trclass="even">
<td>></td>
<tdstyle="text-align: center;">0x3E</td>
<tdstyle="text-align: center;">＞</td>
</tr>
<trclass="odd">
<td>?</td>
<tdstyle="text-align: center;">0x3F</td>
<tdstyle="text-align: center;">？</td>
</tr>
<trclass="even">
<td>:</td>
<tdstyle="text-align: center;">0x3A</td>
<tdstyle="text-align: center;">：</td>
</tr>
<trclass="odd">
<td>|</td>
<tdstyle="text-align: center;">0x7C</td>
<tdstyle="text-align: center;">｜</td>
</tr>
<trclass="even">
<td>"</td>
<tdstyle="text-align: center;">0x22</td>
<tdstyle="text-align: center;">＂</td>
</tr>
</tbody>
</table>
<p>File names can also not start or end with the following characters. These only get replaced if they are the first or last character in the name:</p>
<table>
<thead>
<trclass="header">
<th>Character</th>
<thstyle="text-align: center;">Value</th>
<thstyle="text-align: center;">Replacement</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>SP</td>
<tdstyle="text-align: center;">0x20</td>
<tdstyle="text-align: center;">␠</td>
</tr>
<trclass="even">
<td>.</td>
<tdstyle="text-align: center;">0x2E</td>
<tdstyle="text-align: center;">．</td>
</tr>
</tbody>
</table>
<p>Invalid UTF-8 bytes will also be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as they can't be used in JSON strings.</p>
<h3id="standard-options-8">Standard Options</h3>
<p>Here are the standard options specific to sharefile (Citrix Sharefile).</p>
<p><code>rclone about</code> is not supported by the Citrix ShareFile backend. Backends without this capability cannot determine free space for an rclone mount or use policy <code>mfs</code> (most free space) as a member of an rclone union remote.</p>
<p>See <ahref="https://rclone.org/overview/#optional-features">List of backends that do not support rclone about</a> See <ahref="https://rclone.org/commands/rclone_about/">rclone about</a></p>
<h2id="crypt">Crypt</h2>
<p>Rclone <code>crypt</code> remotes encrypt and decrypt other remotes.</p>
<p>A remote of type <code>crypt</code> does not access a <ahref="https://rclone.org/overview/">storage system</a> directly, but instead wraps another remote, which in turn accesses the storage system. This is similar to how <ahref="https://rclone.org/alias/">alias</a>, <ahref="https://rclone.org/union/">union</a>, <ahref="https://rclone.org/chunker/">chunker</a> and a few others work. It makes the usage very flexible, as you can add a layer, in this case an encryption layer, on top of any other backend, even in multiple layers. Rclone's functionality can be used as with any other remote, for example you can <ahref="https://rclone.org/commands/rclone_mount/">mount</a> a crypt remote.</p>
<p>Accessing a storage system through a crypt remote realizes client-side encryption, which makes it safe to keep your data in a location you do not trust will not get compromised. When working against the <code>crypt</code> remote, rclone will automatically encrypt (before uploading) and decrypt (after downloading) on your local system as needed on the fly, leaving the data encrypted at rest in the wrapped remote. If you access the storage system using an application other than rclone, or access the wrapped remote directly using rclone, there will not be any encryption/decryption: Downloading existing content will just give you the encrypted (scrambled) format, and anything you upload will <em>not</em> become encrypted.</p>
<p>The encryption is a secret-key encryption (also called symmetric key encryption) algorithm, where a password (or pass phrase) is used to generate real encryption key. The password can be supplied by user, or you may chose to let rclone generate one. It will be stored in the configuration file, in a lightly obscured form. If you are in an environment where you are not able to keep your configuration secured, you should add <ahref="https://rclone.org/docs/#configuration-encryption">configuration encryption</a> as protection. As long as you have this configuration file, you will be able to decrypt your data. Without the configuration file, as long as you remember the password (or keep it in a safe place), you can re-create the configuration and gain access to the existing data. You may also configure a corresponding remote in a different installation to access the same data. See below for guidance to <ahref="#changing-password">changing password</a>.</p>
<p>Encryption uses <ahref="https://en.wikipedia.org/wiki/Salt_(cryptography)">cryptographic salt</a>, to permute the encryption key so that the same string may be encrypted in different ways. When configuring the crypt remote it is optional to enter a salt, or to let rclone generate a unique salt. If omitted, rclone uses a built-in unique string. Normally in cryptography, the salt is stored together with the encrypted content, and do not have to be memorized by the user. This is not the case in rclone, because rclone does not store any additional information on the remotes. Use of custom salt is effectively a second password that must be memorized.</p>
<p><ahref="#file-encryption">File content</a> encryption is performed using <ahref="https://godoc.org/golang.org/x/crypto/nacl/secretbox">NaCl SecretBox</a>, based on XSalsa20 cipher and Poly1305 for integrity. <ahref="#name-encryption">Names</a> (file- and directory names) are also encrypted by default, but this has some implications and is therefore possible to turned off.</p>
<h3id="configuration">Configuration</h3>
<p>Here is an example of how to make a remote called <code>secret</code>.</p>
<p>To use <code>crypt</code>, first set up the underlying remote. Follow the <code>rclone config</code> instructions for the specific backend.</p>
<p>Before configuring the crypt remote, check the underlying remote is working. In this example the underlying remote is called <code>remote</code>. We will configure a path <code>path</code> within this remote to contain the encrypted content. Anything inside <code>remote:path</code> will be encrypted and anything outside will not.</p>
<p>Configure <code>crypt</code> using <code>rclone config</code>. In this example the <code>crypt</code> remote is called <code>secret</code>, to differentiate it from the underlying <code>remote</code>.</p>
<p>When you are done you can use the crypt remote named <code>secret</code> just as you would with any other remote, e.g. <code>rclone copy D:\docs secret:\docs</code>, and rclone will encrypt and decrypt as needed on the fly. If you access the wrapped remote <code>remote:path</code> directly you will bypass the encryption, and anything you read will be in encrypted form, and anything you write will be unencrypted. To avoid issues it is best to configure a dedicated path for encrypted content, and access it exclusively through a crypt remote.</p>
NB If filename_encryption is "off" then this option will do nothing.
Enter a boolean value (true or false). Press Enter for the default ("true").
Choose a number from below, or type in your own value
1 / Encrypt directory names.
\ "true"
2 / Don't encrypt directory names, leave them intact.
\ "false"
directory_name_encryption>
Password or pass phrase for encryption.
y) Yes type in my own password
g) Generate random password
y/g> y
Enter the password:
password:
Confirm the password:
password:
Password or pass phrase for salt. Optional but recommended.
Should be different to the previous password.
y) Yes type in my own password
g) Generate random password
n) No leave this optional password blank (default)
y/g/n> g
Password strength in bits.
64 is just about memorable
128 is secure
1024 is the maximum
Bits> 128
Your password is: JAsJvRcgR-_veXNfy_sGmQ
Use this password? Please note that an obscured version of this
password (and not the password itself) will be stored under your
configuration file, so keep this generated password in a safe place.
y) Yes (default)
n) No
y/n>
Edit advanced config? (y/n)
y) Yes
n) No (default)
y/n>
Remote config
--------------------
[secret]
type = crypt
remote = remote:path
password = *** ENCRYPTED ***
password2 = *** ENCRYPTED ***
--------------------
y) Yes this is OK (default)
e) Edit this remote
d) Delete this remote
y/e/d></code></pre>
<p><strong>Important</strong> The crypt password stored in <code>rclone.conf</code> is lightly obscured. That only protects it from cursory inspection. It is not secure unless <ahref="https://rclone.org/docs/#configuration-encryption">configuration encryption</a> of <code>rclone.conf</code> is specified.</p>
<p>A long passphrase is recommended, or <code>rclone config</code> can generate a random one.</p>
<p>The obscured password is created using AES-CTR with a static key. The salt is stored verbatim at the beginning of the obscured password. This static key is shared between all versions of rclone.</p>
<p>If you reconfigure rclone with the same passwords/passphrases elsewhere it will be compatible, but the obscured version will be different due to the different salt.</p>
<p>Rclone does not encrypt</p>
<ul>
<li>file length - this can be calculated within 16 bytes</li>
<li>modification time - used for syncing</li>
</ul>
<h3id="specifying-the-remote-1">Specifying the remote</h3>
<p>When configuring the remote to encrypt/decrypt, you may specify any string that rclone accepts as a source/destination of other commands.</p>
<p>The primary use case is to specify the path into an already configured remote (e.g. <code>remote:path/to/dir</code> or <code>remote:bucket</code>), such that data in a remote untrusted location can be stored encrypted.</p>
<p>You may also specify a local filesystem path, such as <code>/path/to/dir</code> on Linux, <code>C:\path\to\dir</code> on Windows. By creating a crypt remote pointing to such a local filesystem path, you can use rclone as a utility for pure local file encryption, for example to keep encrypted files on a removable USB drive.</p>
<p><strong>Note</strong>: A string which do not contain a <code>:</code> will by rclone be treated as a relative path in the local filesystem. For example, if you enter the name <code>remote</code> without the trailing <code>:</code>, it will be treated as a subdirectory of the current directory with name "remote".</p>
<p>If a path <code>remote:path/to/dir</code> is specified, rclone stores encrypted files in <code>path/to/dir</code> on the remote. With file name encryption, files saved to <code>secret:subdir/subfile</code> are stored in the unencrypted path <code>path/to/dir</code> but the <code>subdir/subpath</code> element is encrypted.</p>
<p>The path you specify does not have to exist, rclone will create it when needed.</p>
<p>If you intend to use the wrapped remote both directly for keeping unencrypted content, as well as through a crypt remote for encrypted content, it is recommended to point the crypt remote to a separate directory within the wrapped remote. If you use a bucket based storage system (e.g. Swift, S3, Google Compute Storage, B2, Hubic) it is generally advisable to wrap the crypt remote around a specific bucket (<code>s3:bucket</code>). If wrapping around the entire root of the storage (<code>s3:</code>), and use the optional file name encryption, rclone will encrypt the bucket name.</p>
<h3id="changing-password">Changing password</h3>
<p>Should the password, or the configuration file containing a lightly obscured form of the password, be compromised, you need to re-encrypt your data with a new password. Since rclone uses secret-key encryption, where the encryption key is generated directly from the password kept on the client, it is not possible to change the password/key of already encrypted content. Just changing the password configured for an existing crypt remote means you will no longer able to decrypt any of the previously encrypted content. The only possibility is to re-upload everything via a crypt remote configured with your new password.</p>
<p>Depending on the size of your data, your bandwith, storage quota etc, there are different approaches you can take: - If you have everything in a different location, for example on your local system, you could remove all of the prior encrypted files, change the password for your configured crypt remote (or delete and re-create the crypt configuration), and then re-upload everything from the alternative location. - If you have enough space on the storage system you can create a new crypt remote pointing to a separate directory on the same backend, and then use rclone to copy everything from the original crypt remote to the new, effectively decrypting everything on the fly using the old password and re-encrypting using the new password. When done, delete the original crypt remote directory and finally the rclone crypt configuration with the old password. All data will be streamed from the storage system and back, so you will get half the bandwith and be charged twice if you have upload and download quota on the storage system.</p>
<p><strong>Note</strong>: A security problem related to the random password generator was fixed in rclone version 1.53.3 (released 2020-11-19). Passwords generated by rclone config in version 1.49.0 (released 2019-08-26) to 1.53.2 (released 2020-10-26) are not considered secure and should be changed. If you made up your own password, or used rclone version older than 1.49.0 or newer than 1.53.2 to generate it, you are <em>not</em> affected by this issue. See <ahref="https://github.com/rclone/rclone/issues/4783">issue #4783</a> for more details, and a tool you can use to check if you are affected.</p>
<h3id="example">Example</h3>
<p>Create the following file structure using "standard" file name encryption.</p>
<p>Without file name encryption <code>.bin</code> extensions are added to underlying names. This prevents the cloud provider attempting to interpret file content.</p>
<pre><code>$ rclone -q ls remote:path
54 file0.txt.bin
57 subdir/file3.txt.bin
56 subdir/file2.txt.bin
58 subdir/subsubdir/file4.txt.bin
55 file1.txt.bin</code></pre>
<h3id="file-name-encryption-modes">File name encryption modes</h3>
<p>Off</p>
<ul>
<li>doesn't hide file names or directory structure</li>
<li>allows for longer file names (~246 characters)</li>
<li>can use sub paths and copy single files</li>
</ul>
<p>Standard</p>
<ul>
<li>file names encrypted</li>
<li>file names can't be as long (~143 characters)</li>
<li>can use sub paths and copy single files</li>
<li>directory structure visible</li>
<li>identical files names will have identical uploaded names</li>
<li>can use shortcuts to shorten the directory recursion</li>
</ul>
<p>Obfuscation</p>
<p>This is a simple "rotate" of the filename, with each file having a rot distance based on the filename. Rclone stores the distance at the beginning of the filename. A file called "hello" may become "53.jgnnq".</p>
<p>Obfuscation is not a strong encryption of filenames, but hinders automated scanning tools picking up on filename patterns. It is an intermediate between "off" and "standard" which allows for longer path segment names.</p>
<p>There is a possibility with some unicode based filenames that the obfuscation is weak and may map lower case characters to upper case equivalents.</p>
<p>Obfuscation cannot be relied upon for strong protection.</p>
<ul>
<li>file names very lightly obfuscated</li>
<li>file names can be longer than standard encryption</li>
<li>can use sub paths and copy single files</li>
<li>directory structure visible</li>
<li>identical files names will have identical uploaded names</li>
</ul>
<p>Cloud storage systems have limits on file name length and total path length which rclone is more likely to breach using "Standard" file name encryption. Where file names are less thn 156 characters in length issues should not be encountered, irrespective of cloud storage provider.</p>
<p>An alternative, future rclone file name encryption mode may tolerate backend provider path length limits.</p>
<h3id="directory-name-encryption">Directory name encryption</h3>
<p>Crypt offers the option of encrypting dir names or leaving them intact. There are two options:</p>
<p>True</p>
<p>Encrypts the whole file path including directory names Example: <code>1/12/123.txt</code> is encrypted to <code>p0e52nreeaj0a5ea7s64m4j72s/l42g6771hnv3an9cgc8cr2n1ng/qgm4avr35m5loi1th53ato71v0</code></p>
<p>False</p>
<p>Only encrypts file names, skips directory names Example: <code>1/12/123.txt</code> is encrypted to <code>1/12/qgm4avr35m5loi1th53ato71v0</code></p>
<h3id="modified-time-and-hashes-3">Modified time and hashes</h3>
<p>Crypt stores modification times using the underlying remote so support depends on that.</p>
<p>Hashes are not stored for crypt. However the data integrity is protected by an extremely strong crypto authenticator.</p>
<p>Use the <code>rclone cryptcheck</code> command to check the integrity of a crypted remote instead of <code>rclone check</code> which can't check the checksums properly.</p>
<h3id="standard-options-9">Standard Options</h3>
<p>Here are the standard options specific to crypt (Encrypt/Decrypt a remote).</p>
<h4id="crypt-remote">--crypt-remote</h4>
<p>Remote to encrypt/decrypt. Normally should contain a ':' and a path, e.g. "myremote:path/to/dir", "myremote:bucket" or maybe "myremote:" (not recommended).</p>
<p>Allow server-side operations (e.g. copy) to work across different crypt configs.</p>
<p>Normally this option is not what you want, but if you have two crypts pointing to the same backend you can use it.</p>
<p>This can be used, for example, to change file name encryption type without re-uploading all the data. Just make two crypt backends pointing to two different directories with the single changed parameter and use rclone move to move the files between the crypt remotes.</p>
<p>For all files listed show how the names encrypt.</p>
<p>If this flag is set then for each file that the remote is asked to list, it will log (at level INFO) a line stating the decrypted file name and the encrypted file name.</p>
<p>This is so you can work out which encrypted names are which decrypted names just in case you need to do something with the encrypted file names, or for debugging purposes.</p>
<p>This decodes the filenames given as arguments returning a list of strings of the decoded results. It will return an error if any of the inputs are invalid.</p>
<h2id="backing-up-a-crypted-remote">Backing up a crypted remote</h2>
<p>If you wish to backup a crypted remote, it is recommended that you use <code>rclone sync</code> on the encrypted files, and make sure the passwords are the same in the new encrypted remote.</p>
<p>This will have the following advantages</p>
<ul>
<li><code>rclone sync</code> will check the checksums while copying</li>
<li>you can use <code>rclone check</code> between the encrypted remotes</li>
<li>you don't decrypt and encrypt unnecessarily</li>
</ul>
<p>For example, let's say you have your original remote at <code>remote:</code> with the encrypted version at <code>eremote:</code> with path <code>remote:crypt</code>. You would then set up the new remote <code>remote2:</code> and then the encrypted version <code>eremote2:</code> with path <code>remote2:crypt</code> using the same passwords as <code>eremote:</code>.</p>
<p>The initial nonce is generated from the operating systems crypto strong random number generator. The nonce is incremented for each chunk read making sure each nonce is unique for each block written. The chance of a nonce being re-used is minuscule. If you wrote an exabyte of data (10¹⁸ bytes) you would have a probability of approximately 2×10⁻³² of re-using a nonce.</p>
<h4id="chunk">Chunk</h4>
<p>Each chunk will contain 64kB of data, except for the last one which may have less data. The data chunk is in standard NaCl SecretBox format. SecretBox uses XSalsa20 and Poly1305 to encrypt and authenticate messages.</p>
<p>Each chunk contains:</p>
<ul>
<li>16 Bytes of Poly1305 authenticator</li>
<li>1 - 65536 bytes XSalsa20 encrypted data</li>
</ul>
<p>64k chunk size was chosen as the best performing chunk size (the authenticator takes too much time below this and the performance drops off due to cache effects above this). Note that these chunks are buffered in memory so they can't be too big.</p>
<p>This uses a 32 byte (256 bit key) key derived from the user password.</p>
<h4id="examples">Examples</h4>
<p>1 byte file will encrypt to</p>
<ul>
<li>32 bytes header</li>
<li>17 bytes data chunk</li>
</ul>
<p>49 bytes total</p>
<p>1MB (1048576 bytes) file will encrypt to</p>
<ul>
<li>32 bytes header</li>
<li>16 chunks of 65568 bytes</li>
</ul>
<p>1049120 bytes total (a 0.05% overhead). This is the overhead for big files.</p>
<h3id="name-encryption">Name encryption</h3>
<p>File names are encrypted segment by segment - the path is broken up into <code>/</code> separated strings and these are encrypted individually.</p>
<p>File segments are padded using PKCS#7 to a multiple of 16 bytes before encryption.</p>
<p>They are then encrypted with EME using AES with 256 bit key. EME (ECB-Mix-ECB) is a wide-block encryption mode presented in the 2003 paper "A Parallelizable Enciphering Mode" by Halevi and Rogaway.</p>
<p>This makes for deterministic encryption which is what we want - the same filename must encrypt to the same thing otherwise we can't find it on the cloud storage system.</p>
<p>This means that</p>
<ul>
<li>filenames with the same name will encrypt the same</li>
<li>filenames which start the same won't have a common prefix</li>
</ul>
<p>This uses a 32 byte key (256 bits) and a 16 byte (128 bits) IV both of which are derived from the user password.</p>
<p>After encryption they are written out using a modified version of standard <code>base32</code> encoding as described in RFC4648. The standard encoding is modified in two ways:</p>
<ul>
<li>it becomes lower case (no-one likes upper case filenames!)</li>
<li>we strip the padding character <code>=</code></li>
</ul>
<p><code>base32</code> is used rather than the more efficient <code>base64</code> so rclone can be used on case insensitive remotes (e.g. Windows, Amazon Drive).</p>
<h3id="key-derivation">Key derivation</h3>
<p>Rclone uses <code>scrypt</code> with parameters <code>N=16384, r=8, p=1</code> with an optional user supplied salt (password2) to derive the 32+32+16 = 80 bytes of key material required. If the user doesn't supply a salt then rclone uses an internal one.</p>
<p><code>scrypt</code> makes it impractical to mount a dictionary attack on rclone encrypted data. For full protection against this you should always use a salt.</p>
<p>This remote is currently <strong>experimental</strong>. Things may break and data may be lost. Anything you do with this remote is at your own risk. Please understand the risks associated with using experimental code and don't use this remote in critical applications.</p>
<p>The <code>Compress</code> remote adds compression to another remote. It is best used with remotes containing many large compressible files.</p>
<p>To use this remote, all you need to do is specify another remote and a compression mode to use:</p>
Enter a string value. Press Enter for the default ("").
remote> remote_to_press:subdir
Compression mode.
Enter a string value. Press Enter for the default ("gzip").
Choose a number from below, or type in your own value
1 / Gzip compression balanced for speed and compression strength.
\ "gzip"
compression_mode> gzip
Edit advanced config? (y/n)
y) Yes
n) No (default)
y/n> n
Remote config
--------------------
[compress]
type = compress
remote = remote_to_press:subdir
compression_mode = gzip
--------------------
y) Yes this is OK (default)
e) Edit this remote
d) Delete this remote
y/e/d> y</code></pre>
<h3id="compression-modes">Compression Modes</h3>
<p>Currently only gzip compression is supported, it provides a decent balance between speed and strength and is well supported by other application. Compression strength can further be configured via an advanced setting where 0 is no compression and 9 is strongest compression.</p>
<h4id="filetype">Filetype</h4>
<p>If you open a remote wrapped by press, you will see that there are many files with an extension corresponding to the compression algorithm you chose. These files are standard files that can be opened by various archive programs, but they have some hidden metadata that allows them to be used by rclone. While you may download and decompress these files at will, do <strong>not</strong> manually delete or rename files. Files without correct metadata files will not be recognized by rclone.</p>
<h3id="file-names">File names</h3>
<p>The compressed files will be named <code>*.###########.gz</code> where <code>*</code> is the base file and the <code>#</code> part is base64 encoded size of the uncompressed file. The file names should not be changed by anything other than the rclone compression backend.</p>
<h3id="standard-options-10">Standard Options</h3>
<p>Here are the standard options specific to compress (Compress a remote).</p>
<h4id="compress-remote">--compress-remote</h4>
<p>Remote to compress.</p>
<ul>
<li>Config: remote</li>
<li>Env Var: RCLONE_COMPRESS_REMOTE</li>
<li>Type: string</li>
<li>Default: ""</li>
</ul>
<h4id="compress-mode">--compress-mode</h4>
<p>Compression mode.</p>
<ul>
<li>Config: mode</li>
<li>Env Var: RCLONE_COMPRESS_MODE</li>
<li>Type: string</li>
<li>Default: "gzip"</li>
<li>Examples:
<ul>
<li>"gzip"
<ul>
<li>Standard gzip compression with fastest parameters.</li>
</ul></li>
</ul></li>
</ul>
<h3id="advanced-options-9">Advanced Options</h3>
<p>Here are the advanced options specific to compress (Compress a remote).</p>
<h4id="compress-level">--compress-level</h4>
<p>GZIP compression level (-2 to 9).</p>
<pre><code> Generally -1 (default, equivalent to 5) is recommended.
Levels 1 to 9 increase compressiong at the cost of speed.. Going past 6
generally offers very little return.
Level -2 uses Huffmann encoding only. Only use if you now what you
<p>Some remotes don't allow the upload of files with unknown size. In this case the compressed file will need to be cached to determine it's size.</p>
<pre><code> Files smaller than this limit will be cached in RAM, file larger than
this limit will be cached on disk</code></pre>
<ul>
<li>Config: ram_cache_limit</li>
<li>Env Var: RCLONE_COMPRESS_RAM_CACHE_LIMIT</li>
<li>Type: SizeSuffix</li>
<li>Default: 20M</li>
</ul>
<h2id="dropbox">Dropbox</h2>
<p>Paths are specified as <code>remote:path</code></p>
<p>Dropbox paths may be as deep as required, e.g. <code>remote:directory/subdirectory</code>.</p>
<p>The initial setup for dropbox involves getting a token from Dropbox which you need to do in your browser. <code>rclone config</code> walks you through it.</p>
<p>Here is an example of how to make a remote called <code>remote</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>n) New remote
d) Delete remote
q) Quit config
e/n/d/q> n
name> remote
Type of storage to configure.
Choose a number from below, or type in your own value
<h3id="dropbox-for-business">Dropbox for business</h3>
<p>Rclone supports Dropbox for business and Team Folders.</p>
<p>When using Dropbox for business <code>remote:</code> and <code>remote:path/to/file</code> will refer to your personal folder.</p>
<p>If you wish to see Team Folders you must use a leading <code>/</code> in the path, so <code>rclone lsd remote:/</code> will refer to the root and show you all Team Folders and your User Folder.</p>
<p>You can then use team folders like this <code>remote:/TeamFolder</code> and <code>remote:/TeamFolder/path/to/file</code>.</p>
<p>A leading <code>/</code> for a Dropbox personal account will do nothing, but it will take an extra HTTP transaction so it should be avoided.</p>
<h3id="modified-time-and-hashes-4">Modified time and Hashes</h3>
<p>Dropbox supports modified times, but the only way to set a modification time is to re-upload the file.</p>
<p>This means that if you uploaded your data with an older version of rclone which didn't support the v2 API and modified times, rclone will decide to upload all your old data to fix the modification times. If you don't want this to happen use <code>--size-only</code> or <code>--checksum</code> flag to stop it.</p>
<p>Dropbox supports <ahref="https://www.dropbox.com/developers/reference/content-hash">its own hash type</a> which is checked for all transfers.</p>
<p>Any files larger than this will be uploaded in chunks of this size.</p>
<p>Note that chunks are buffered in memory (one at a time) so rclone can deal with retries. Setting this larger will increase the speed slightly (at most 10% for 128MB in tests) at the cost of using more memory. It can be set smaller if you are tight on memory.</p>
<p>Note that if you want to use impersonate, you should make sure this flag is set when running "rclone config" as this will cause rclone to request the "members.read" scope which it won't normally. This is needed to lookup a members email address into the internal ID that dropbox uses in the API.</p>
<p>Using the "members.read" scope will require a Dropbox Team Admin to approve during the OAuth flow.</p>
<p>You will have to use your own App (setting your own client_id and client_secret) to use this option as currently rclone's default set of permissions doesn't include "members.read". This can be added once v1.55 or later is in use everywhere.</p>
<p>Instructs rclone to work on individual shared files.</p>
<p>In this mode rclone's features are extremely limited - only list (ls, lsl, etc.) operations and read operations (e.g. downloading) are supported in this mode. All other operations will be disabled.</p>
<p>Instructs rclone to work on shared folders.</p>
<p>When this flag is used with no path only the List operation is supported and all available shared folders will be listed. If you specify a path the first part will be interpreted as the name of shared folder. Rclone will then try to mount this shared to the root namespace. On success shared folder rclone proceeds normally. The shared folder is now pretty much a normal folder and all normal operations are supported.</p>
<p>Note that we don't unmount the shared folder afterwards so the --dropbox-shared-folders can be omitted after the first use of a particular shared folder.</p>
<ul>
<li>Config: shared_folders</li>
<li>Env Var: RCLONE_DROPBOX_SHARED_FOLDERS</li>
<li>Type: bool</li>
<li>Default: false</li>
</ul>
<h4id="dropbox-encoding">--dropbox-encoding</h4>
<p>This sets the encoding for the backend.</p>
<p>See: the <ahref="https://rclone.org/overview/#encoding">encoding section in the overview</a> for more info.</p>
<p>Note that Dropbox is case insensitive so you can't have a file called "Hello.doc" and one called "hello.doc".</p>
<p>There are some file names such as <code>thumbs.db</code> which Dropbox can't store. There is a full list of them in the <ahref="https://www.dropbox.com/en/help/145">"Ignored Files" section of this document</a>. Rclone will issue an error message <code>File name disallowed - not uploading</code> if it attempts to upload one of those file names, but the sync won't fail.</p>
<p>Some errors may occur if you try to sync copyright-protected files because Dropbox has its own <ahref="https://techcrunch.com/2014/03/30/how-dropbox-knows-when-youre-sharing-copyrighted-stuff-without-actually-looking-at-your-stuff/">copyright detector</a> that prevents this sort of file being downloaded. This will return the error <code>ERROR : /path/to/your/file: Failed to copy: failed to open source object: path/restricted_content/.</code></p>
<p>If you have more than 10,000 files in a directory then <code>rclone purge dropbox:dir</code> will return the error <code>Failed to purge: There are too many files involved in this operation</code>. As a work-around do an <code>rclone delete dropbox:dir</code> followed by an <code>rclone rmdir dropbox:dir</code>.</p>
<h3id="get-your-own-dropbox-app-id">Get your own Dropbox App ID</h3>
<p>When you use rclone with Dropbox in its default configuration you are using rclone's App ID. This is shared between all the rclone users.</p>
<p>Here is how to create your own Dropbox App ID for rclone:</p>
<oltype="1">
<li><p>Log into the <ahref="https://www.dropbox.com/developers/apps/create">Dropbox App console</a> with your Dropbox Account (It need not to be the same account as the Dropbox you want to access)</p></li>
<li><p>Choose an API => Usually this should be <code>Dropbox API</code></p></li>
<li><p>Choose the type of access you want to use =><code>Full Dropbox</code> or <code>App Folder</code></p></li>
<li><p>Name your App. The app name is global, so you can't use <code>rclone</code> for example</p></li>
<li><p>Click the button <code>Create App</code></p></li>
<li><p>Fill <code>Redirect URIs</code> as <code>http://localhost:53682/</code></p></li>
<li><p>Find the <code>App key</code> and <code>App secret</code> Use these values in rclone config to add a new remote or edit an existing remote.</p></li>
<p>This backend supports <ahref="https://storagemadeeasy.com/about/">Storage Made Easy's Enterprise File Fabric™</a> which provides a software solution to integrate and unify File and Object Storage accessible through a global file system.</p>
<p>The initial setup for the Enterprise File Fabric backend involves getting a token from the the Enterprise File Fabric which you need to do in your browser. <code>rclone config</code> walks you through it.</p>
<p>Here is an example of how to make a remote called <code>remote</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> remote
Type of storage to configure.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
[snip]
XX / Enterprise File Fabric
\ "filefabric"
[snip]
Storage> filefabric
** See help for filefabric backend at: https://rclone.org/filefabric/ **
<h3id="modified-time-and-hashes-5">Modified time and hashes</h3>
<p>The Enterprise File Fabric allows modification times to be set on files accurate to 1 second. These will be used to detect whether objects need syncing or not.</p>
<p>The Enterprise File Fabric does not support any data hashes at this time.</p>
<p>The <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a> will be replaced.</p>
<p>Invalid UTF-8 bytes will also be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as they can't be used in JSON strings.</p>
<h3id="empty-files">Empty files</h3>
<p>Empty files aren't supported by the Enterprise File Fabric. Rclone will therefore upload an empty file as a single space with a mime type of <code>application/vnd.rclone.empty.file</code> and files with that mime type are treated as empty.</p>
<h3id="root-folder-id-1">Root folder ID</h3>
<p>You can set the <code>root_folder_id</code> for rclone. This is the directory (identified by its <code>Folder ID</code>) that rclone considers to be the root of your Enterprise File Fabric.</p>
<p>Normally you will leave this blank and rclone will determine the correct root to use itself.</p>
<p>However you can set this to restrict rclone to a specific folder hierarchy.</p>
<p>In order to do this you will have to find the <code>Folder ID</code> of the directory you wish rclone to display. These aren't displayed in the web interface, but you can use <code>rclone lsf</code> to find them, for example</p>
<p>A Permanent Authentication Token can be created in the Enterprise File Fabric, on the users Dashboard under Security, there is an entry you'll see called "My Authentication Tokens". Click the Manage button to create one.</p>
<p>These tokens are normally valid for several years.</p>
<p>For more info see: https://docs.storagemadeeasy.com/organisationcloud/api-tokens</p>
<p>See: the <ahref="https://rclone.org/overview/#encoding">encoding section in the overview</a> for more info.</p>
<ul>
<li>Config: encoding</li>
<li>Env Var: RCLONE_FILEFABRIC_ENCODING</li>
<li>Type: MultiEncoder</li>
<li>Default: Slash,Del,Ctl,InvalidUtf8,Dot</li>
</ul>
<h2id="ftp">FTP</h2>
<p>FTP is the File Transfer Protocol. Rclone FTP support is provided using the <ahref="https://godoc.org/github.com/jlaffaye/ftp">github.com/jlaffaye/ftp</a> package.</p>
<p><ahref="#Limitations">Limitations of Rclone's FTP backend</a></p>
<p>Paths are specified as <code>remote:path</code>. If the path does not begin with a <code>/</code> it is relative to the home directory of the user. An empty path <code>remote:</code> refers to the user's home directory.</p>
<p>To create an FTP configuration named <code>remote</code>, run</p>
<pre><code>rclone config</code></pre>
<p>Rclone config guides you through an interactive setup process. A minimal rclone FTP remote definition only requires host, username and password. For an anonymous FTP server, use <code>anonymous</code> as username and your email address as password.</p>
<pre><code>No remotes found - make a new one
n) New remote
r) Rename remote
c) Copy remote
s) Set configuration password
q) Quit config
n/r/c/s/q> n
name> remote
Type of storage to configure.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
[snip]
XX / FTP Connection
\ "ftp"
[snip]
Storage> ftp
** See help for ftp backend at: https://rclone.org/ftp/ **
<p>Rlone FTP supports implicit FTP over TLS servers (FTPS). This has to be enabled in the FTP backend config for the remote, or with <ahref="#ftp-tls"><code>--ftp-tls</code></a>. The default FTPS port is <code>990</code>, not <code>21</code> and can be set with <ahref="#ftp-port"><code>--ftp-port</code></a>.</p>
<p>Here are the standard options specific to ftp (FTP Connection).</p>
<h4id="ftp-host">--ftp-host</h4>
<p>FTP host to connect to</p>
<ul>
<li>Config: host</li>
<li>Env Var: RCLONE_FTP_HOST</li>
<li>Type: string</li>
<li>Default: ""</li>
<li>Examples:
<ul>
<li>"ftp.example.com"
<ul>
<li>Connect to ftp.example.com</li>
</ul></li>
</ul></li>
</ul>
<h4id="ftp-user">--ftp-user</h4>
<p>FTP username, leave blank for current username, $USER</p>
<ul>
<li>Config: user</li>
<li>Env Var: RCLONE_FTP_USER</li>
<li>Type: string</li>
<li>Default: ""</li>
</ul>
<h4id="ftp-port">--ftp-port</h4>
<p>FTP port, leave blank to use default (21)</p>
<ul>
<li>Config: port</li>
<li>Env Var: RCLONE_FTP_PORT</li>
<li>Type: string</li>
<li>Default: ""</li>
</ul>
<h4id="ftp-pass">--ftp-pass</h4>
<p>FTP password</p>
<p><strong>NB</strong> Input to this must be obscured - see <ahref="https://rclone.org/commands/rclone_obscure/">rclone obscure</a>.</p>
<ul>
<li>Config: pass</li>
<li>Env Var: RCLONE_FTP_PASS</li>
<li>Type: string</li>
<li>Default: ""</li>
</ul>
<h4id="ftp-tls">--ftp-tls</h4>
<p>Use Implicit FTPS (FTP over TLS) When using implicit FTP over TLS the client connects using TLS right from the start which breaks compatibility with non-TLS-aware servers. This is usually served over port 990 rather than port 21. Cannot be used in combination with explicit FTP.</p>
<ul>
<li>Config: tls</li>
<li>Env Var: RCLONE_FTP_TLS</li>
<li>Type: bool</li>
<li>Default: false</li>
</ul>
<h4id="ftp-explicit-tls">--ftp-explicit-tls</h4>
<p>Use Explicit FTPS (FTP over TLS) When using explicit FTP over TLS the client explicitly requests security from the server in order to upgrade a plain text connection to an encrypted one. Cannot be used in combination with implicit FTP.</p>
<ul>
<li>Config: explicit_tls</li>
<li>Env Var: RCLONE_FTP_EXPLICIT_TLS</li>
<li>Type: bool</li>
<li>Default: false</li>
</ul>
<h3id="advanced-options-12">Advanced Options</h3>
<p>Here are the advanced options specific to ftp (FTP Connection).</p>
<h4id="ftp-concurrency">--ftp-concurrency</h4>
<p>Maximum number of FTP simultaneous connections, 0 for unlimited</p>
<p>See: the <ahref="https://rclone.org/overview/#encoding">encoding section in the overview</a> for more info.</p>
<ul>
<li>Config: encoding</li>
<li>Env Var: RCLONE_FTP_ENCODING</li>
<li>Type: MultiEncoder</li>
<li>Default: Slash,Del,Ctl,RightSpace,Dot</li>
</ul>
<h3id="limitations-10">Limitations</h3>
<p>Modified times are not supported. Times you see on the FTP server through rclone are those of upload.</p>
<p>Rclone's FTP backend does not support any checksums but can compare file sizes.</p>
<p><code>rclone about</code> is not supported by the FTP backend. Backends without this capability cannot determine free space for an rclone mount or use policy <code>mfs</code> (most free space) as a member of an rclone union remote.</p>
<p>See <ahref="https://rclone.org/overview/#optional-features">List of backends that do not support rclone about</a> See <ahref="https://rclone.org/commands/rclone_about/">rclone about</a></p>
<p>The implementation of : <code>--dump headers</code>, <code>--dump bodies</code>, <code>--dump auth</code> for debugging isn't the same as for rclone HTTP based backends - it has less fine grained control.</p>
<p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a> the following characters are also replaced:</p>
<p>File names cannot end with the following characters. Repacement is limited to the last character in a file name:</p>
<table>
<thead>
<trclass="header">
<th>Character</th>
<thstyle="text-align: center;">Value</th>
<thstyle="text-align: center;">Replacement</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>SP</td>
<tdstyle="text-align: center;">0x20</td>
<tdstyle="text-align: center;">␠</td>
</tr>
</tbody>
</table>
<p>Not all FTP servers can have all characters in file names, for example:</p>
<p>Paths are specified as <code>remote:bucket</code> (or <code>remote:</code> for the <code>lsd</code> command.) You may put subdirectories in too, e.g. <code>remote:bucket/path/to/dir</code>.</p>
<p>The initial setup for google cloud storage involves getting a token from Google Cloud Storage which you need to do in your browser. <code>rclone config</code> walks you through it.</p>
<p>Here is an example of how to make a remote called <code>remote</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>n) New remote
d) Delete remote
q) Quit config
e/n/d/q> n
name> remote
Type of storage to configure.
Choose a number from below, or type in your own value
[snip]
XX / Google Cloud Storage (this is not Google Drive)
\ "google cloud storage"
[snip]
Storage> google cloud storage
Google Application Client Id - leave blank normally.
client_id>
Google Application Client Secret - leave blank normally.
client_secret>
Project number optional - needed only for list/create/delete buckets - see your developer console.
project_number> 12345678
Service Account Credentials JSON file path - needed only if you want use SA instead of interactive login.
service_account_file>
Access Control List for new objects.
Choose a number from below, or type in your own value
1 / Object owner gets OWNER access, and all Authenticated Users get READER access.
\ "authenticatedRead"
2 / Object owner gets OWNER access, and project team owners get OWNER access.
\ "bucketOwnerFullControl"
3 / Object owner gets OWNER access, and project team owners get READER access.
\ "bucketOwnerRead"
4 / Object owner gets OWNER access [default if left blank].
\ "private"
5 / Object owner gets OWNER access, and project team members get access according to their roles.
\ "projectPrivate"
6 / Object owner gets OWNER access, and all Users get READER access.
\ "publicRead"
object_acl> 4
Access Control List for new buckets.
Choose a number from below, or type in your own value
1 / Project team owners get OWNER access, and all Authenticated Users get READER access.
\ "authenticatedRead"
2 / Project team owners get OWNER access [default if left blank].
\ "private"
3 / Project team members get access according to their roles.
\ "projectPrivate"
4 / Project team owners get OWNER access, and all Users get READER access.
\ "publicRead"
5 / Project team owners get OWNER access, and all Users get WRITER access.
\ "publicReadWrite"
bucket_acl> 2
Location for the newly created buckets.
Choose a number from below, or type in your own value
1 / Empty for default location (US).
\ ""
2 / Multi-regional location for Asia.
\ "asia"
3 / Multi-regional location for Europe.
\ "eu"
4 / Multi-regional location for United States.
\ "us"
5 / Taiwan.
\ "asia-east1"
6 / Tokyo.
\ "asia-northeast1"
7 / Singapore.
\ "asia-southeast1"
8 / Sydney.
\ "australia-southeast1"
9 / Belgium.
\ "europe-west1"
10 / London.
\ "europe-west2"
11 / Iowa.
\ "us-central1"
12 / South Carolina.
\ "us-east1"
13 / Northern Virginia.
\ "us-east4"
14 / Oregon.
\ "us-west1"
location> 12
The storage class to use when storing objects in Google Cloud Storage.
Choose a number from below, or type in your own value
1 / Default
\ ""
2 / Multi-regional storage class
\ "MULTI_REGIONAL"
3 / Regional storage class
\ "REGIONAL"
4 / Nearline storage class
\ "NEARLINE"
5 / Coldline storage class
\ "COLDLINE"
6 / Durable reduced availability storage class
\ "DURABLE_REDUCED_AVAILABILITY"
storage_class> 5
Remote config
Use auto config?
* Say Y if not sure
* Say N if you are working on a remote or headless machine or Y didn't work
y) Yes
n) No
y/n> y
If your browser doesn't open automatically go to the following link: http://127.0.0.1:53682/auth
<p>Note that rclone runs a webserver on your local machine to collect the token as returned from Google if you use auto config mode. This only runs from the moment it opens your browser to the moment you get back the verification code. This is on <code>http://127.0.0.1:53682/</code> and this it may require you to unblock it temporarily if you are running a host firewall, or use manual mode.</p>
<p>This remote is called <code>remote</code> and can now be used like this</p>
<p>You can set up rclone with Google Cloud Storage in an unattended mode, i.e. not tied to a specific end-user Google account. This is useful when you want to synchronise files onto machines that don't have actively logged-in users, for example build machines.</p>
<p>To get credentials for Google Cloud Platform <ahref="https://cloud.google.com/iam/docs/service-accounts">IAM Service Accounts</a>, please head to the <ahref="https://console.cloud.google.com/permissions/serviceaccounts">Service Account</a> section of the Google Developer Console. Service Accounts behave just like normal <code>User</code> permissions in <ahref="https://cloud.google.com/storage/docs/access-control">Google Cloud Storage ACLs</a>, so you can limit their access (e.g. make them read only). After creating an account, a JSON file containing the Service Account's credentials will be downloaded onto your machines. These credentials are what rclone will use for authentication.</p>
<p>To use a Service Account instead of OAuth2 token flow, enter the path to your Service Account credentials at the <code>service_account_file</code> prompt and rclone won't use the browser based authentication flow. If you'd rather stuff the contents of the credentials file into the rclone config file, you can set <code>service_account_credentials</code> with the actual contents of the file instead, or set the equivalent environment variable.</p>
<h3id="anonymous-access">Anonymous Access</h3>
<p>For downloads of objects that permit public access you can configure rclone to use anonymous access by setting <code>anonymous</code> to <code>true</code>. With unauthorized access you can't write or create files but only read or list those buckets and objects that have public read access.</p>
<p>If no other source of credentials is provided, rclone will fall back to <ahref="https://cloud.google.com/video-intelligence/docs/common/auth#authenticating_with_application_default_credentials">Application Default Credentials</a> this is useful both when you already have configured authentication for your developer account, or in production when running on a google compute host. Note that if running in docker, you may need to run additional commands on your google compute machine - <ahref="https://cloud.google.com/container-registry/docs/advanced-authentication#gcloud_as_a_docker_credential_helper">see this page</a>.</p>
<p>Note that in the case application default credentials are used, there is no need to explicitly configure a project number.</p>
<h3id="fast-list-2">--fast-list</h3>
<p>This remote supports <code>--fast-list</code> which allows you to use fewer transactions in exchange for more memory. See the <ahref="https://rclone.org/docs/#fast-list">rclone docs</a> for more details.</p>
<p>You can set custom upload headers with the <code>--header-upload</code> flag. Google Cloud Storage supports the headers as described in the <ahref="https://cloud.google.com/storage/docs/gsutil/addlhelp/WorkingWithObjectMetadata">working with metadata documentation</a></p>
<p>Note that the last of these is for setting custom metadata in the form <code>--header-upload "x-goog-meta-key: value"</code></p>
<h3id="modified-time-3">Modified time</h3>
<p>Google google cloud storage stores md5sums natively and rclone stores modification times as metadata on the object, under the "mtime" key in RFC3339 format accurate to 1ns.</p>
<p>Here are the advanced options specific to google cloud storage (Google Cloud Storage (this is not Google Drive)).</p>
<h4id="gcs-token">--gcs-token</h4>
<p>OAuth Access Token as a JSON blob.</p>
<ul>
<li>Config: token</li>
<li>Env Var: RCLONE_GCS_TOKEN</li>
<li>Type: string</li>
<li>Default: ""</li>
</ul>
<h4id="gcs-auth-url">--gcs-auth-url</h4>
<p>Auth server URL. Leave blank to use the provider defaults.</p>
<ul>
<li>Config: auth_url</li>
<li>Env Var: RCLONE_GCS_AUTH_URL</li>
<li>Type: string</li>
<li>Default: ""</li>
</ul>
<h4id="gcs-token-url">--gcs-token-url</h4>
<p>Token server url. Leave blank to use the provider defaults.</p>
<ul>
<li>Config: token_url</li>
<li>Env Var: RCLONE_GCS_TOKEN_URL</li>
<li>Type: string</li>
<li>Default: ""</li>
</ul>
<h4id="gcs-encoding">--gcs-encoding</h4>
<p>This sets the encoding for the backend.</p>
<p>See: the <ahref="https://rclone.org/overview/#encoding">encoding section in the overview</a> for more info.</p>
<ul>
<li>Config: encoding</li>
<li>Env Var: RCLONE_GCS_ENCODING</li>
<li>Type: MultiEncoder</li>
<li>Default: Slash,CrLf,InvalidUtf8,Dot</li>
</ul>
<h3id="limitations-11">Limitations</h3>
<p><code>rclone about</code> is not supported by the Google Cloud Storage backend. Backends without this capability cannot determine free space for an rclone mount or use policy <code>mfs</code> (most free space) as a member of an rclone union remote.</p>
<p>See <ahref="https://rclone.org/overview/#optional-features">List of backends that do not support rclone about</a> See <ahref="https://rclone.org/commands/rclone_about/">rclone about</a></p>
<h2id="google-drive">Google Drive</h2>
<p>Paths are specified as <code>drive:path</code></p>
<p>Drive paths may be as deep as required, e.g. <code>drive:directory/subdirectory</code>.</p>
<p>The initial setup for drive involves getting a token from Google drive which you need to do in your browser. <code>rclone config</code> walks you through it.</p>
<p>Here is an example of how to make a remote called <code>remote</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>No remotes found - make a new one
n) New remote
r) Rename remote
c) Copy remote
s) Set configuration password
q) Quit config
n/r/c/s/q> n
name> remote
Type of storage to configure.
Choose a number from below, or type in your own value
[snip]
XX / Google Drive
\ "drive"
[snip]
Storage> drive
Google Application Client Id - leave blank normally.
client_id>
Google Application Client Secret - leave blank normally.
<p>Note that rclone runs a webserver on your local machine to collect the token as returned from Google if you use auto config mode. This only runs from the moment it opens your browser to the moment you get back the verification code. This is on <code>http://127.0.0.1:53682/</code> and this it may require you to unblock it temporarily if you are running a host firewall, or use manual mode.</p>
<p>You can then use it like this,</p>
<p>List directories in top level of your drive</p>
<pre><code>rclone lsd remote:</code></pre>
<p>List all the files in your drive</p>
<pre><code>rclone ls remote:</code></pre>
<p>To copy a local directory to a drive directory called backup</p>
<p>Rclone allows you to select which scope you would like for rclone to use. This changes what type of token is granted to rclone. <ahref="https://developers.google.com/drive/v3/web/about-auth">The scopes are defined here</a>.</p>
<p>The scope are</p>
<h4id="drive">drive</h4>
<p>This is the default scope and allows full access to all files, except for the Application Data Folder (see below).</p>
<p>Choose this one if you aren't sure.</p>
<h4id="drive.readonly">drive.readonly</h4>
<p>This allows read only access to all files. Files may be listed and downloaded but not uploaded, renamed or deleted.</p>
<h4id="drive.file">drive.file</h4>
<p>With this scope rclone can read/view/modify only those files and folders it creates.</p>
<p>So if you uploaded files to drive via the web interface (or any other means) they will not be visible to rclone.</p>
<p>This can be useful if you are using rclone to backup data and you want to be sure confidential data on your drive is not visible to rclone.</p>
<p>Files created with this scope are visible in the web interface.</p>
<h4id="drive.appfolder">drive.appfolder</h4>
<p>This gives rclone its own private area to store files. Rclone will not be able to see any other files on your drive and you won't be able to see rclone's files from the web interface either.</p>
<p>This allows read only access to file names only. It does not allow rclone to download or upload data, or rename or delete files or directories.</p>
<h3id="root-folder-id-2">Root folder ID</h3>
<p>You can set the <code>root_folder_id</code> for rclone. This is the directory (identified by its <code>Folder ID</code>) that rclone considers to be the root of your drive.</p>
<p>Normally you will leave this blank and rclone will determine the correct root to use itself.</p>
<p>However you can set this to restrict rclone to a specific folder hierarchy or to access data within the "Computers" tab on the drive web interface (where files from Google's Backup and Sync desktop program go).</p>
<p>In order to do this you will have to find the <code>Folder ID</code> of the directory you wish rclone to display. This will be the last segment of the URL when you open the relevant folder in the drive web interface.</p>
<p>So if the folder you want rclone to use has a URL which looks like <code>https://drive.google.com/drive/folders/1XyfxxxxxxxxxxxxxxxxxxxxxxxxxKHCh</code> in the browser, then you use <code>1XyfxxxxxxxxxxxxxxxxxxxxxxxxxKHCh</code> as the <code>root_folder_id</code> in the config.</p>
<p><strong>NB</strong> folders under the "Computers" tab seem to be read only (drive gives a 500 error) when using rclone.</p>
<p>There doesn't appear to be an API to discover the folder IDs of the "Computers" tab - please contact us if you know otherwise!</p>
<p>Note also that rclone can't access any data under the "Backups" tab on the google drive web interface yet.</p>
<p>You can set up rclone with Google Drive in an unattended mode, i.e. not tied to a specific end-user Google account. This is useful when you want to synchronise files onto machines that don't have actively logged-in users, for example build machines.</p>
<p>To use a Service Account instead of OAuth2 token flow, enter the path to your Service Account credentials at the <code>service_account_file</code> prompt during <code>rclone config</code> and rclone won't use the browser based authentication flow. If you'd rather stuff the contents of the credentials file into the rclone config file, you can set <code>service_account_credentials</code> with the actual contents of the file instead, or set the equivalent environment variable.</p>
<h4id="use-case---google-appsg-suite-account-and-individual-drive">Use case - Google Apps/G-suite account and individual Drive</h4>
<p>Let's say that you are the administrator of a Google Apps (old) or G-suite account. The goal is to store data on an individual's Drive account, who IS a member of the domain. We'll call the domain <strong>example.com</strong>, and the user <strong>foo@example.com</strong>.</p>
<p>There's a few steps we need to go through to accomplish this:</p>
<h5id="create-a-service-account-for-example.com">1. Create a service account for example.com</h5>
<ul>
<li>To create a service account and obtain its credentials, go to the <ahref="https://console.developers.google.com">Google Developer Console</a>.</li>
<li>You must have a project - create one if you don't.</li>
<li>Then go to "IAM & admin" -> "Service Accounts".</li>
<li>Use the "Create Credentials" button. Fill in "Service account name" with something that identifies your client. "Role" can be empty.</li>
<li>Tick "Furnish a new private key" - select "Key type JSON".</li>
<li>Tick "Enable G Suite Domain-wide Delegation". This option makes "impersonation" possible, as documented here: <ahref="https://developers.google.com/identity/protocols/OAuth2ServiceAccount#delegatingauthority">Delegating domain-wide authority to the service account</a></li>
<li>These credentials are what rclone will use for authentication. If you ever need to remove access, press the "Delete service account key" button.</li>
</ul>
<h5id="allowing-api-access-to-example.com-google-drive">2. Allowing API access to example.com Google Drive</h5>
<ul>
<li>Go to example.com's admin console</li>
<li>Go into "Security" (or use the search bar)</li>
<li>Select "Show more" and then "Advanced settings"</li>
<li>Select "Manage API client access" in the "Authentication" section</li>
<li>In the "Client Name" field enter the service account's "Client ID" - this can be found in the Developer Console under "IAM & Admin" -> "Service Accounts", then "View Client ID" for the newly created service account. It is a ~21 character numerical string.</li>
<li>In the next field, "One or More API Scopes", enter <code>https://www.googleapis.com/auth/drive</code> to grant access to Google Drive specifically.</li>
</ul>
<h5id="configure-rclone-assuming-a-new-install">3. Configure rclone, assuming a new install</h5>
<pre><code>rclone config
n/s/q> n # New
name>gdrive # Gdrive is an example name
Storage> # Select the number shown for Google Drive
client_id> # Can be left blank
client_secret> # Can be left blank
scope> # Select your scope, 1 for example
root_folder_id> # Can be left blank
service_account_file> /home/foo/myJSONfile.json # This is where the JSON file goes!
y/n> # Auto config, y
</code></pre>
<h5id="verify-that-its-working">4. Verify that it's working</h5>
<li><code>--drive-impersonate foo@example.com</code> - this is what does the magic, pretending to be user foo.</li>
<li><code>lsf</code> - list files in a parsing friendly way</li>
<li><code>gdrive:backup</code> - use the remote called gdrive, work in the folder named backup.</li>
</ul></li>
</ul>
<p>Note: in case you configured a specific root folder on gdrive and rclone is unable to access the contents of that folder when using <code>--drive-impersonate</code>, do this instead: - in the gdrive web interface, share your root folder with the user/email of the new Service Account you created/selected at step #1 - use rclone without specifying the <code>--drive-impersonate</code> option, like this: <code>rclone -v foo@example.com lsf gdrive:backup</code></p>
<p>If you want to configure the remote to point to a Google Shared Drive (previously known as Team Drives) then answer <code>y</code> to the question <code>Configure this as a Shared Drive (Team Drive)?</code>.</p>
<p>This will fetch the list of Shared Drives from google and allow you to configure which one you want to use. You can also type in a Shared Drive ID if you prefer.</p>
<p>This remote supports <code>--fast-list</code> which allows you to use fewer transactions in exchange for more memory. See the <ahref="https://rclone.org/docs/#fast-list">rclone docs</a> for more details.</p>
<p>It does this by combining multiple <code>list</code> calls into a single API request.</p>
<p>This works by combining many <code>'%s' in parents</code> filters into one expression. To list the contents of directories a, b and c, the following requests will be send by the regular <code>List</code> function:</p>
<pre><code>trashed=false and 'a' in parents
trashed=false and 'b' in parents
trashed=false and 'c' in parents</code></pre>
<p>These can now be combined into a single request:</p>
<pre><code>trashed=false and ('a' in parents or 'b' in parents or 'c' in parents)</code></pre>
<p>The implementation of <code>ListR</code> will put up to 50 <code>parents</code> filters into one request. It will use the <code>--checkers</code> value to specify the number of requests to run in parallel.</p>
<p>In tests, these batch requests were up to 20x faster than the regular method. Running the following command against different sized folders gives:</p>
<p>Only Invalid UTF-8 bytes will be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as they can't be used in JSON strings.</p>
<p>In contrast to other backends, <code>/</code> can also be used in names and <code>.</code> or <code>..</code> are valid names.</p>
<h3id="revisions">Revisions</h3>
<p>Google drive stores revisions of files. When you upload a change to an existing file to google drive using rclone it will create a new revision of that file.</p>
<p>Revisions follow the standard google policy which at time of writing was</p>
<ul>
<li>They are deleted after 30 days or 100 revisions (whatever comes first).</li>
<li>They do not count towards a user storage quota.</li>
</ul>
<h3id="deleting-files-2">Deleting files</h3>
<p>By default rclone will send all files to the trash when deleting files. If deleting them permanently is required then use the <code>--drive-use-trash=false</code> flag, or set the equivalent environment variable.</p>
<h3id="shortcuts">Shortcuts</h3>
<p>In March 2020 Google introduced a new feature in Google Drive called <ahref="https://support.google.com/drive/answer/9700156">drive shortcuts</a> (<ahref="https://developers.google.com/drive/api/v3/shortcuts">API</a>). These will (by September 2020) <ahref="https://cloud.google.com/blog/products/g-suite/simplifying-google-drives-folder-structure-and-sharing-models">replace the ability for files or folders to be in multiple folders at once</a>.</p>
<p>Shortcuts are files that link to other files on Google Drive somewhat like a symlink in unix, except they point to the underlying file data (e.g. the inode in unix terms) so they don't break if the source is renamed or moved about.</p>
<p>Be default rclone treats these as follows.</p>
<p>For shortcuts pointing to files:</p>
<ul>
<li>When listing a file shortcut appears as the destination file.</li>
<li>When downloading the contents of the destination file is downloaded.</li>
<li>When updating shortcut file with a non shortcut file, the shortcut is removed then a new file is uploaded in place of the shortcut.</li>
<li>When server-side moving (renaming) the shortcut is renamed, not the destination file.</li>
<li>When server-side copying the shortcut is copied, not the contents of the shortcut.</li>
<li>When deleting the shortcut is deleted not the linked file.</li>
<li>When setting the modification time, the modification time of the linked file will be set.</li>
</ul>
<p>For shortcuts pointing to folders:</p>
<ul>
<li>When listing the shortcut appears as a folder and that folder will contain the contents of the linked folder appear (including any sub folders)</li>
<li>When downloading the contents of the linked folder and sub contents are downloaded</li>
<li>When uploading to a shortcut folder the file will be placed in the linked folder</li>
<li>When server-side moving (renaming) the shortcut is renamed, not the destination folder</li>
<li>When server-side copying the contents of the linked folder is copied, not the shortcut.</li>
<li>When deleting with <code>rclone rmdir</code> or <code>rclone purge</code> the shortcut is deleted not the linked folder.</li>
<li><strong>NB</strong> When deleting with <code>rclone remove</code> or <code>rclone mount</code> the contents of the linked folder will be deleted.</li>
</ul>
<p>The <ahref="https://rclone.org/commands/rclone_backend/">rclone backend</a> command can be used to create shortcuts.</p>
<p>Shortcuts can be completely ignored with the <code>--drive-skip-shortcuts</code> flag or the corresponding <code>skip_shortcuts</code> configuration setting.</p>
<h3id="emptying-trash">Emptying trash</h3>
<p>If you wish to empty your trash you can use the <code>rclone cleanup remote:</code> command which will permanently delete all your trashed files. This command does not take any path arguments.</p>
<p>Note that Google Drive takes some time (minutes to days) to empty the trash even though the command returns within a few seconds. No output is echoed, so there will be no confirmation even using -v or -vv.</p>
<h3id="quota-information">Quota information</h3>
<p>To view your current quota you can use the <code>rclone about remote:</code> command which will display your usage limit (quota), the usage in Google Drive, the size of all files in the Trash and the space used by other Google services such as Gmail. This command does not take any path arguments.</p>
<h4id="importexport-of-google-documents">Import/Export of google documents</h4>
<p>Google documents can be exported from and uploaded to Google Drive.</p>
<p>When rclone downloads a Google doc it chooses a format to download depending upon the <code>--drive-export-formats</code> setting. By default the export formats are <code>docx,xlsx,pptx,svg</code> which are a sensible default for an editable document.</p>
<p>When choosing a format, rclone runs down the list provided in order and chooses the first file format the doc can be exported as from the list. If the file can't be exported to a format on the formats list, then rclone will choose a format from the default list.</p>
<p>If you prefer an archive copy then you might use <code>--drive-export-formats pdf</code>, or if you prefer openoffice/libreoffice formats you might use <code>--drive-export-formats ods,odt,odp</code>.</p>
<p>Note that rclone adds the extension to the google doc, so if it is called <code>My Spreadsheet</code> on google docs, it will be exported as <code>My Spreadsheet.xlsx</code> or <code>My Spreadsheet.pdf</code> etc.</p>
<p>When importing files into Google Drive, rclone will convert all files with an extension in <code>--drive-import-formats</code> to their associated document type. rclone will not convert any files by default, since the conversion is lossy process.</p>
<p>The conversion must result in a file with the same extension when the <code>--drive-export-formats</code> rules are applied to the uploaded document.</p>
<p>Here are some examples for allowed and prohibited conversions.</p>
<table>
<thead>
<trclass="header">
<th>export-formats</th>
<th>import-formats</th>
<th>Upload Ext</th>
<th>Document Ext</th>
<th>Allowed</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>odt</td>
<td>odt</td>
<td>odt</td>
<td>odt</td>
<td>Yes</td>
</tr>
<trclass="even">
<td>odt</td>
<td>docx,odt</td>
<td>odt</td>
<td>odt</td>
<td>Yes</td>
</tr>
<trclass="odd">
<td></td>
<td>docx</td>
<td>docx</td>
<td>docx</td>
<td>Yes</td>
</tr>
<trclass="even">
<td></td>
<td>odt</td>
<td>odt</td>
<td>docx</td>
<td>No</td>
</tr>
<trclass="odd">
<td>odt,docx</td>
<td>docx,odt</td>
<td>docx</td>
<td>odt</td>
<td>No</td>
</tr>
<trclass="even">
<td>docx,odt</td>
<td>docx,odt</td>
<td>docx</td>
<td>docx</td>
<td>Yes</td>
</tr>
<trclass="odd">
<td>docx,odt</td>
<td>docx,odt</td>
<td>odt</td>
<td>docx</td>
<td>No</td>
</tr>
</tbody>
</table>
<p>This limitation can be disabled by specifying <code>--drive-allow-import-name-change</code>. When using this flag, rclone can convert multiple files types resulting in the same document type at once, e.g. with <code>--drive-import-formats docx,odt,txt</code>, all files having these extension would result in a document represented as a docx file. This brings the additional risk of overwriting a document, if multiple files have the same stem. Many rclone operations will not handle this name change in any way. They assume an equal name when copying files and might copy the file again or delete them when the name changes.</p>
<p>Here are the possible export extensions with their corresponding mime types. Most of these can also be used for importing, but there more that are not listed here. Some of these additional ones might only be available when the operating system provides the correct MIME type entries.</p>
<p>This list can be changed by Google Drive at any time and might not represent the currently available conversions.</p>
<p>Google documents can also be exported as link files. These files will open a browser window for the Google Docs website of that document when opened. The link file extension has to be specified as a <code>--drive-export-formats</code> parameter. They will match all available Google Documents.</p>
<table>
<thead>
<trclass="header">
<th>Extension</th>
<th>Description</th>
<th>OS Support</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>desktop</td>
<td>freedesktop.org specified desktop entry</td>
<td>Linux</td>
</tr>
<trclass="even">
<td>link.html</td>
<td>An HTML Document with a redirect</td>
<td>All</td>
</tr>
<trclass="odd">
<td>url</td>
<td>INI style link file</td>
<td>macOS, Windows</td>
</tr>
<trclass="even">
<td>webloc</td>
<td>macOS specific XML format</td>
<td>macOS</td>
</tr>
</tbody>
</table>
<h3id="standard-options-15">Standard Options</h3>
<p>Here are the standard options specific to drive (Google Drive).</p>
<h4id="drive-client-id">--drive-client-id</h4>
<p>Google Application Client Id Setting your own is recommended. See https://rclone.org/drive/#making-your-own-client-id for how to create your own. If you leave this blank, it will use an internal key which is low performance.</p>
<p>Only consider files owned by the authenticated user.</p>
<ul>
<li>Config: auth_owner_only</li>
<li>Env Var: RCLONE_DRIVE_AUTH_OWNER_ONLY</li>
<li>Type: bool</li>
<li>Default: false</li>
</ul>
<h4id="drive-use-trash">--drive-use-trash</h4>
<p>Send files to the trash instead of deleting permanently. Defaults to true, namely sending files to the trash. Use <code>--drive-use-trash=false</code> to delete files permanently instead.</p>
<ul>
<li>Config: use_trash</li>
<li>Env Var: RCLONE_DRIVE_USE_TRASH</li>
<li>Type: bool</li>
<li>Default: true</li>
</ul>
<h4id="drive-skip-gdocs">--drive-skip-gdocs</h4>
<p>Skip google documents in all listings. If given, gdocs practically become invisible to rclone.</p>
<p>Instructs rclone to operate on your "Shared with me" folder (where Google Drive lets you access the files and folders others have shared with you).</p>
<p>This works both with the "list" (lsd, lsl, etc.) and the "copy" commands (copy, sync, etc.), and with all other commands too.</p>
<p>Use file created date instead of modified date.,</p>
<p>Useful when downloading data and you want the creation date used in place of the last modified date.</p>
<p><strong>WARNING</strong>: This flag may have some unexpected consequences.</p>
<p>When uploading to your drive all files will be overwritten unless they haven't been modified since their creation. And the inverse will occur while downloading. This side effect can be avoided by using the "--checksum" flag.</p>
<p>This feature was implemented to retain photos capture date as recorded by google photos. You will first need to check the "Create a Google Photos folder" option in your google drive settings. You can then copy or move the photos locally and use the date the image was taken (created) set as the modification date.</p>
<p>Set to allow files which return cannotDownloadAbusiveFile to be downloaded.</p>
<p>If downloading a file returns the error "This file has been identified as malware or spam and cannot be downloaded" with the error code "cannotDownloadAbusiveFile" then supply this flag to rclone to indicate you acknowledge the risks of downloading the file and rclone will download it anyway.</p>
<p>Show sizes as storage quota usage, not actual size.</p>
<p>Show the size of a file as the storage quota used. This is the current version plus any older versions that have been set to keep forever.</p>
<p><strong>WARNING</strong>: This flag may have some unexpected consequences.</p>
<p>It is not recommended to set this flag in your config - the recommended usage is using the flag form --drive-size-as-quota when doing rclone ls/lsl/lsf/lsjson/etc only.</p>
<p>If you do use this flag for syncing (not recommended) then you will need to use --ignore size also.</p>
<p>Allow server-side operations (e.g. copy) to work across different drive configs.</p>
<p>This can be useful if you wish to do a server-side copy between two different Google drives. Note that this isn't enabled by default because it isn't easy to tell if it will work between any two configurations.</p>
<p>There is currently an unsolved issue with the google drive backend and HTTP/2. HTTP/2 is therefore disabled by default for the drive backend but can be re-enabled here. When the issue is solved this flag will be removed.</p>
<p>At the time of writing it is only possible to upload 750GB of data to Google Drive a day (this is an undocumented limit). When this limit is reached Google Drive produces a slightly different error message. When this flag is set it causes these errors to be fatal. These will stop the in-progress sync.</p>
<p>Note that this detection is relying on error message strings which Google don't document so it may break in the future.</p>
<p>At the time of writing it is only possible to download 10TB of data from Google Drive a day (this is an undocumented limit). When this limit is reached Google Drive produces a slightly different error message. When this flag is set it causes these errors to be fatal. These will stop the in-progress sync.</p>
<p>Note that this detection is relying on error message strings which Google don't document so it may break in the future.</p>
<p>Normally rclone dereferences shortcut files making them appear as if they are the original file (see <ahref="#shortcuts">the shortcuts section</a>). If this flag is set then rclone will ignore shortcut files completely.</p>
<ul>
<li>Config: skip_shortcuts</li>
<li>Env Var: RCLONE_DRIVE_SKIP_SHORTCUTS</li>
<li>Type: bool</li>
<li>Default: false</li>
</ul>
<h4id="drive-encoding">--drive-encoding</h4>
<p>This sets the encoding for the backend.</p>
<p>See: the <ahref="https://rclone.org/overview/#encoding">encoding section in the overview</a> for more info.</p>
<ul>
<li>Config: encoding</li>
<li>Env Var: RCLONE_DRIVE_ENCODING</li>
<li>Type: MultiEncoder</li>
<li>Default: InvalidUtf8</li>
</ul>
<h3id="backend-commands-3">Backend commands</h3>
<p>Here are the commands specific to the drive backend.</p>
<p>In the first example this creates a shortcut from the "source_item" which can be a file or a directory to the "destination_shortcut". The "source_item" and the "destination_shortcut" should be relative paths from "drive:"</p>
<p>In the second example this creates a shortcut from the "source_item" relative to "drive:" to the "destination_shortcut" relative to "drive2:". This may fail with a permission error if the user authenticated with "drive2:" can't read files from "drive:".</p>
<p>Options:</p>
<ul>
<li>"target": optional target remote for the shortcut destination</li>
<p>It copies the drive file with ID given to the path (an rclone path which will be passed internally to rclone copyto). The ID and path pairs can be repeated.</p>
<p>The path should end with a / to indicate copy the file as named to this directory. If it doesn't end with a / then the last path component will be used as the file name.</p>
<p>If the destination is a drive backend then server-side copying will be attempted if possible.</p>
<p>Use the -i flag to see what would be copied before copying.</p>
<h3id="limitations-12">Limitations</h3>
<p>Drive has quite a lot of rate limiting. This causes rclone to be limited to transferring about 2 files per second only. Individual files may be transferred much faster at 100s of MBytes/s but lots of small files can take a long time.</p>
<p>Server side copies are also subject to a separate rate limit. If you see User rate limit exceeded errors, wait at least 24 hours and retry. You can disable server-side copies with <code>--disable copy</code> to download and upload the files if you prefer.</p>
<h4id="limitations-of-google-docs">Limitations of Google Docs</h4>
<p>Google docs will appear as size -1 in <code>rclone ls</code> and as size 0 in anything which uses the VFS layer, e.g. <code>rclone mount</code>, <code>rclone serve</code>.</p>
<p>This is because rclone can't find out the size of the Google docs without downloading them.</p>
<p>Google docs will transfer correctly with <code>rclone sync</code>, <code>rclone copy</code> etc as rclone knows to ignore the size when doing the transfer.</p>
<p>However an unfortunate consequence of this is that you may not be able to download Google docs using <code>rclone mount</code>. If it doesn't work you will get a 0 sized file. If you try again the doc may gain its correct size and be downloadable. Whether it will work on not depends on the application accessing the mount and the OS you are running - experiment to find out if it does work for you!</p>
<h3id="duplicated-files-1">Duplicated files</h3>
<p>Sometimes, for no reason I've been able to track down, drive will duplicate a file that rclone uploads. Drive unlike all the other remotes can have duplicated files.</p>
<p>Duplicated files cause problems with the syncing and you will see messages in the log about duplicates.</p>
<p>Use <code>rclone dedupe</code> to fix duplicated files.</p>
<p>Note that this isn't just a problem with rclone, even Google Photos on Android duplicates files on drive sometimes.</p>
<h3id="rclone-appears-to-be-re-copying-files-it-shouldnt">Rclone appears to be re-copying files it shouldn't</h3>
<p>The most likely cause of this is the duplicated file issue above - run <code>rclone dedupe</code> and check your logs for duplicate object or directory messages.</p>
<p>This can also be caused by a delay/caching on google drive's end when comparing directory listings. Specifically with team drives used in combination with --fast-list. Files that were uploaded recently may not appear on the directory list sent to rclone when using --fast-list.</p>
<p>Waiting a moderate period of time between attempts (estimated to be approximately 1 hour) and/or not using --fast-list both seem to be effective in preventing the problem.</p>
<h3id="making-your-own-client_id">Making your own client_id</h3>
<p>When you use rclone with Google drive in its default configuration you are using rclone's client_id. This is shared between all the rclone users. There is a global rate limit on the number of queries per second that each client_id can do set by Google. rclone already has a high quota and I will continue to make sure it is high enough by contacting Google.</p>
<p>It is strongly recommended to use your own client ID as the default rclone ID is heavily used. If you have multiple services running, it is recommended to use an API key for each service. The default Google quota is 10 transactions per second so it is recommended to stay under that number as if you use more than that, it will cause rclone to rate limit and make things slower.</p>
<p>Here is how to create your own Google Drive client ID for rclone:</p>
<oltype="1">
<li><p>Log into the <ahref="https://console.developers.google.com/">Google API Console</a> with your Google account. It doesn't matter what Google account you use. (It need not be the same account as the Google Drive you want to access)</p></li>
<li><p>Select a project or create a new project.</p></li>
<li><p>Under "ENABLE APIS AND SERVICES" search for "Drive", and enable the "Google Drive API".</p></li>
<li><p>Click "Credentials" in the left-side panel (not "Create credentials", which opens the wizard), then "Create credentials"</p></li>
<li><p>If you already configured an "Oauth Consent Screen", then skip to the next step; if not, click on "CONFIGURE CONSENT SCREEN" button (near the top right corner of the right panel), then select "External" and click on "CREATE"; on the next screen, enter an "Application name" ("rclone" is OK) then click on "Save" (all other data is optional). Click again on "Credentials" on the left panel to go back to the "Credentials" screen.</p></li>
</ol>
<p>(PS: if you are a GSuite user, you could also select "Internal" instead of "External" above, but this has not been tested/documented so far).</p>
<olstart="6"type="1">
<li><p>Click on the "+ CREATE CREDENTIALS" button at the top of the screen, then select "OAuth client ID".</p></li>
<li><p>Choose an application type of "Desktop app" if you using a Google account or "Other" if you using a GSuite account and click "Create". (the default name is fine)</p></li>
<li><p>It will show you a client ID and client secret. Use these values in rclone config to add a new remote or edit an existing remote.</p></li>
</ol>
<p>Be aware that, due to the "enhanced security" recently introduced by Google, you are theoretically expected to "submit your app for verification" and then wait a few weeks(!) for their response; in practice, you can go right ahead and use the client ID and client secret with rclone, the only issue will be a very scary confirmation screen shown when you connect via your browser for rclone to be able to get its token-id (but as this only happens during the remote configuration, it's not such a big deal).</p>
<p>(Thanks to <spanclass="citation"data-cites="balazer">@balazer</span> on github for these instructions.)</p>
<p>Sometimes, creation of an OAuth consent in Google API Console fails due to an error message “The request failed because changes to one of the field of the resource is not supported”. As a convenient workaround, the necessary Google Drive API key can be created on the <ahref="https://developers.google.com/drive/api/v3/quickstart/python">Python Quickstart</a> page. Just push the Enable the Drive API button to receive the Client ID and Secret. Note that it will automatically create a new project in the API Console.</p>
<h2id="google-photos">Google Photos</h2>
<p>The rclone backend for <ahref="https://www.google.com/photos/about/">Google Photos</a> is a specialized backend for transferring photos and videos to and from Google Photos.</p>
<p><strong>NB</strong> The Google Photos API which rclone uses has quite a few limitations, so please read the <ahref="#limitations">limitations section</a> carefully to make sure it is suitable for your use.</p>
<h2id="configuring-google-photos">Configuring Google Photos</h2>
<p>The initial setup for google cloud storage involves getting a token from Google Photos which you need to do in your browser. <code>rclone config</code> walks you through it.</p>
<p>Here is an example of how to make a remote called <code>remote</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> remote
Type of storage to configure.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
[snip]
XX / Google Photos
\ "google photos"
[snip]
Storage> google photos
** See help for google photos backend at: https://rclone.org/googlephotos/ **
<p>Note that rclone runs a webserver on your local machine to collect the token as returned from Google if you use auto config mode. This only runs from the moment it opens your browser to the moment you get back the verification code. This is on <code>http://127.0.0.1:53682/</code> and this may require you to unblock it temporarily if you are running a host firewall, or use manual mode.</p>
<p>This remote is called <code>remote</code> and can now be used like this</p>
<p>As Google Photos is not a general purpose cloud storage system the backend is laid out to help you navigate it.</p>
<p>The directories under <code>media</code> show different ways of categorizing the media. Each file will appear multiple times. So if you want to make a backup of your google photos you might choose to backup <code>remote:media/by-month</code>. (<strong>NB</strong><code>remote:media/by-day</code> is rather slow at the moment so avoid for syncing.)</p>
<p>Note that all your photos and videos will appear somewhere under <code>media</code>, but they may not appear under <code>album</code> unless you've put them into albums.</p>
<pre><code>/
- upload
- file1.jpg
- file2.jpg
- ...
- media
- all
- file1.jpg
- file2.jpg
- ...
- by-year
- 2000
- file1.jpg
- ...
- 2001
- file2.jpg
- ...
- ...
- by-month
- 2000
- 2000-01
- file1.jpg
- ...
- 2000-02
- file2.jpg
- ...
- ...
- by-day
- 2000
- 2000-01-01
- file1.jpg
- ...
- 2000-01-02
- file2.jpg
- ...
- ...
- album
- album name
- album name/sub
- shared-album
- album name
- album name/sub
- feature
- favorites
- file1.jpg
- file2.jpg</code></pre>
<p>There are two writable parts of the tree, the <code>upload</code> directory and sub directories of the <code>album</code> directory.</p>
<p>The <code>upload</code> directory is for uploading files you don't want to put into albums. This will be empty to start with and will contain the files you've uploaded for one rclone session only, becoming empty again when you restart rclone. The use case for this would be if you have a load of files you just want to once off dump into Google Photos. For repeated syncing, uploading to <code>album</code> will work better.</p>
<p>Directories within the <code>album</code> directory are also writeable and you may create new directories (albums) under <code>album</code>. If you copy files with a directory hierarchy in there then rclone will create albums with the <code>/</code> character in them. For example if you do</p>
<p>Then rclone will create the following albums with the following files in</p>
<ul>
<li>images
<ul>
<li>file1.jpg</li>
</ul></li>
<li>images/dir
<ul>
<li>file2.jpg</li>
</ul></li>
<li>images/dir2/dir3
<ul>
<li>file3.jpg</li>
</ul></li>
</ul>
<p>This means that you can use the <code>album</code> path pretty much like a normal filesystem and it is a good target for repeated syncing.</p>
<p>The <code>shared-album</code> directory shows albums shared with you or by you. This is similar to the Sharing tab in the Google Photos web interface.</p>
<h2id="limitations-13">Limitations</h2>
<p>Only images and videos can be uploaded. If you attempt to upload non videos or images or formats that Google Photos doesn't understand, rclone will upload the file, then Google Photos will give an error when it is put turned into a media item.</p>
<p>Note that all media items uploaded to Google Photos through the API are stored in full resolution at "original quality" and <strong>will</strong> count towards your storage quota in your Google Account. The API does <strong>not</strong> offer a way to upload in "high quality" mode..</p>
<p><code>rclone about</code> is not supported by the Google Photos backend. Backends without this capability cannot determine free space for an rclone mount or use policy <code>mfs</code> (most free space) as a member of an rclone union remote.</p>
<p>See <ahref="https://rclone.org/overview/#optional-features">List of backends that do not support rclone about</a> See <ahref="https://rclone.org/commands/rclone_about/">rclone about</a></p>
<p>When Images are downloaded this strips EXIF location (according to the docs and my tests). This is a limitation of the Google Photos API and is covered by <ahref="https://issuetracker.google.com/issues/112096115">bug #112096115</a>.</p>
<p><strong>The current google API does not allow photos to be downloaded at original resolution. This is very important if you are, for example, relying on "Google Photos" as a backup of your photos. You will not be able to use rclone to redownload original images. You could use 'google takeout' to recover the original photos as a last resort</strong></p>
<p>When videos are downloaded they are downloaded in a really compressed version of the video compared to downloading it via the Google Photos web interface. This is covered by <ahref="https://issuetracker.google.com/issues/113672044">bug #113672044</a>.</p>
<h3id="duplicates">Duplicates</h3>
<p>If a file name is duplicated in a directory then rclone will add the file ID into its name. So two files called <code>file.jpg</code> would then appear as <code>file {123456}.jpg</code> and <code>file {ABCDEF}.jpg</code> (the actual IDs are a lot longer alas!).</p>
<p>If you upload the same image (with the same binary data) twice then Google Photos will deduplicate it. However it will retain the filename from the first upload which may confuse rclone. For example if you uploaded an image to <code>upload</code> then uploaded the same image to <code>album/my_album</code> the filename of the image in <code>album/my_album</code> will be what it was uploaded with initially, not what you uploaded it with to <code>album</code>. In practise this shouldn't cause too many problems.</p>
<h3id="modified-time-5">Modified time</h3>
<p>The date shown of media in Google Photos is the creation date as determined by the EXIF information, or the upload date if that is not known.</p>
<p>This is not changeable by rclone and is not the modification date of the media on local disk. This means that rclone cannot use the dates from Google Photos for syncing purposes.</p>
<h3id="size">Size</h3>
<p>The Google Photos API does not return the size of media. This means that when syncing to Google Photos, rclone can only do a file existence check.</p>
<p>It is possible to read the size of the media, but this needs an extra HTTP HEAD request per media item so is <strong>very slow</strong> and uses up a lot of transactions. This can be enabled with the <code>--gphotos-read-size</code> option or the <code>read_size = true</code> config parameter.</p>
<p>If you want to use the backend with <code>rclone mount</code> you may need to enable this flag (depending on your OS and application using the photos) otherwise you may not be able to read media off the mount. You'll need to experiment to see if it works for you without the flag.</p>
<h3id="albums">Albums</h3>
<p>Rclone can only upload files to albums it created. This is a <ahref="https://developers.google.com/photos/library/guides/manage-albums">limitation of the Google Photos API</a>.</p>
<p>Rclone can remove files it uploaded from albums it created only.</p>
<h3id="deleting-files-3">Deleting files</h3>
<p>Rclone can remove files from albums it created, but note that the Google Photos API does not allow media to be deleted permanently so this media will still remain. See <ahref="https://issuetracker.google.com/issues/109759781">bug #109759781</a>.</p>
<p>Rclone cannot delete files anywhere except under <code>album</code>.</p>
<h3id="deleting-albums">Deleting albums</h3>
<p>The Google Photos API does not support deleting albums - see <ahref="https://issuetracker.google.com/issues/135714733">bug #135714733</a>.</p>
<h3id="standard-options-16">Standard Options</h3>
<p>Here are the standard options specific to google photos (Google Photos).</p>
<p>Normally rclone does not read the size of media items since this takes another transaction. This isn't necessary for syncing. However rclone mount needs to know the size of files in advance of reading them, so setting this flag when using rclone mount is recommended if you want to read the media.</p>
<p>By default rclone does not request archived media. Thus, when syncing, archived media is not visible in directory listings or transferred.</p>
<p>Note that media in albums is always visible and synced, no matter their archive status.</p>
<p>With this flag, archived media are always visible in directory listings and transferred.</p>
<p>Without this flag, archived media will not be visible in directory listings and won't be transferred.</p>
<ul>
<li>Config: include_archived</li>
<li>Env Var: RCLONE_GPHOTOS_INCLUDE_ARCHIVED</li>
<li>Type: bool</li>
<li>Default: false</li>
</ul>
<h2id="hdfs">HDFS</h2>
<p><ahref="https://hadoop.apache.org/docs/stable/hadoop-project-dist/hadoop-hdfs/HdfsDesign.html">HDFS</a> is a distributed file-system, part of the <ahref="https://hadoop.apache.org/">Apache Hadoop</a> framework.</p>
<p>Paths are specified as <code>remote:</code> or <code>remote:path/to/dir</code>.</p>
<p>Here is an example of how to make a remote called <code>remote</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> remote
Type of storage to configure.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
[skip]
XX / Hadoop distributed file system
\ "hdfs"
[skip]
Storage> hdfs
** See help for hdfs backend at: https://rclone.org/hdfs/ **
hadoop name node and port
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
1 / Connect to host namenode at port 8020
\ "namenode:8020"
namenode> namenode.hadoop:8020
hadoop user name
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
<h3id="setting-up-your-own-hdfs-instance-for-testing">Setting up your own HDFS instance for testing</h3>
<p>You may start with a <ahref="https://hadoop.apache.org/docs/stable/hadoop-project-dist/hadoop-common/SingleCluster.html">manual setup</a> or use the docker image from the tests:</p>
<p><strong>NB</strong> it need few seconds to startup.</p>
<p>For this docker image the remote needs to be configured like this:</p>
<pre><code>[remote]
type = hdfs
namenode = 127.0.0.1:8020
username = root</code></pre>
<p>You can stop this image with <code>docker kill rclone-hdfs</code> (<strong>NB</strong> it does not use volumes, so all data uploaded will be lost.)</p>
<h3id="modified-time-6">Modified time</h3>
<p>Time accurate to 1 second is stored.</p>
<h3id="checksum">Checksum</h3>
<p>No checksums are implemented.</p>
<h3id="usage-information">Usage information</h3>
<p>You can use the <code>rclone about remote:</code> command which will display filesystem size and current usage.</p>
<p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a> the following characters are also replaced:</p>
<p>Kerberos data transfer protection: authentication|integrity|privacy</p>
<p>Specifies whether or not authentication, data signature integrity checks, and wire encryption is required when communicating the the datanodes. Possible values are 'authentication', 'integrity' and 'privacy'. Used only with KERBEROS enabled.</p>
<p>The HTTP remote is a read only remote for reading files of a webserver. The webserver should provide file listings which rclone will read and turn into a remote. This has been tested with common webservers such as Apache/Nginx/Caddy and will likely work with file listings from most web servers. (If it doesn't then please file an issue, or send a pull request!)</p>
<p>Use this if your target website does not use / on the end of directories.</p>
<p>A / on the end of a path is how rclone normally tells the difference between files and directories. If this flag is set, then rclone will treat all files with Content-Type: text/html as directories and read URLs from them rather than downloading them.</p>
<p>Note that this may cause rclone to confuse genuine HTML files with directories.</p>
<p>If your site is being very slow to load then you can try this option. Normally rclone does a HEAD request for each potential file in a directory listing to:</p>
<ul>
<li>find its size</li>
<li>check it really exists</li>
<li>check to see if it is a directory</li>
</ul>
<p>If you set this option, rclone will not do the HEAD request. This will mean</p>
<p><code>rclone about</code> is not supported by the HTTP backend. Backends without this capability cannot determine free space for an rclone mount or use policy <code>mfs</code> (most free space) as a member of an rclone union remote.</p>
<p>See <ahref="https://rclone.org/overview/#optional-features">List of backends that do not support rclone about</a> See <ahref="https://rclone.org/commands/rclone_about/">rclone about</a></p>
<p>Paths are specified as <code>remote:container</code> (or <code>remote:</code> for the <code>lsd</code> command.) You may put subdirectories in too, e.g. <code>remote:container/path/to/dir</code>.</p>
<p>The initial setup for Hubic involves getting a token from Hubic which you need to do in your browser. <code>rclone config</code> walks you through it.</p>
<p>Note that rclone runs a webserver on your local machine to collect the token as returned from Hubic. This only runs from the moment it opens your browser to the moment you get back the verification code. This is on <code>http://127.0.0.1:53682/</code> and this it may require you to unblock it temporarily if you are running a host firewall.</p>
<p>If you want the directory to be visible in the official <em>Hubic browser</em>, you need to copy your files to the <code>default</code> directory</p>
<p>This remote supports <code>--fast-list</code> which allows you to use fewer transactions in exchange for more memory. See the <ahref="https://rclone.org/docs/#fast-list">rclone docs</a> for more details.</p>
<p>The Swift API doesn't return a correct MD5SUM for segmented files (Dynamic or Static Large Objects) so rclone won't check or use the MD5SUM for these.</p>
<p>Jottacloud is a cloud storage service provider from a Norwegian company, using its own datacenters in Norway.</p>
<p>In addition to the official service at <ahref="https://www.jottacloud.com/">jottacloud.com</a>, there are also several whitelabel versions which should work with this backend.</p>
<p>To configure Jottacloud you will need to generate a personal security token in the Jottacloud web interface. You will the option to do in your <ahref="https://www.jottacloud.com/web/secure">account security settings</a> (for whitelabel version you need to find this page in its web interface). Note that the web interface may refer to this token as a JottaCli token.</p>
<p>If you are using one of the whitelabel versions (Elgiganten, Com Hem Cloud) you may not have the option to generate a CLI token. In this case you'll have to use the legacy authentication. To to this select yes when the setup asks for legacy authentication and enter your username and password. The rest of the setup is identical to the default setup.</p>
<h3id="telia-cloud-setup">Telia Cloud Setup</h3>
<p>Similar to other whitelabel versions Telia Cloud doesn't offer the option of creating a CLI token, and additionally uses a separate authentication flow where the username is generated internally. To setup rclone to use Telia Cloud, choose Telia Cloud authentication in the setup. The rest of the setup is identical to the default setup.</p>
<p>The official Jottacloud client registers a device for each computer you install it on, and then creates a mountpoint for each folder you select for Backup. The web interface uses a special device called Jotta for the Archive and Sync mountpoints. In most cases you'll want to use the Jotta/Archive device/mountpoint, however if you want to access files uploaded by any of the official clients rclone provides the option to select other devices and mountpoints during config.</p>
<p>The built-in Jotta device may also contain several other mountpoints, such as: Latest, Links, Shared and Trash. These are special mountpoints with a different internal representation than the "regular" mountpoints. Rclone will only to a very limited degree support them. Generally you should avoid these, unless you know what you are doing.</p>
<p>This remote supports <code>--fast-list</code> which allows you to use fewer transactions in exchange for more memory. See the <ahref="https://rclone.org/docs/#fast-list">rclone docs</a> for more details.</p>
<p>Note that the implementation in Jottacloud always uses only a single API request to get the entire list, so for large folders this could lead to long wait time before the first results are shown.</p>
<p>Jottacloud allows modification times to be set on objects accurate to 1 second. These will be used to detect whether objects need syncing or not.</p>
<p>Jottacloud supports MD5 type hashes, so you can use the <code>--checksum</code> flag.</p>
<p>Note that Jottacloud requires the MD5 hash before upload so if the source does not have an MD5 checksum then the file will be cached temporarily on disk (wherever the <code>TMPDIR</code> environment variable points to) before it is uploaded. Small files will be cached in memory - see the <ahref="#jottacloud-md5-memory-limit">--jottacloud-md5-memory-limit</a> flag. When uploading from local disk the source checksum is always available, so this does not apply. Starting with rclone version 1.52 the same is true for crypted remotes (in older versions the crypt backend would not calculate hashes for uploads from local disk, so the Jottacloud backend had to do it as described above).</p>
<p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a> the following characters are also replaced:</p>
<p>By default rclone will send all files to the trash when deleting files. They will be permanently deleted automatically after 30 days. You may bypass the trash and permanently delete files immediately by using the <ahref="#jottacloud-hard-delete">--jottacloud-hard-delete</a> flag, or set the equivalent environment variable. Emptying the trash is supported by the <ahref="https://rclone.org/commands/rclone_cleanup/">cleanup</a> command.</p>
<p>Jottacloud supports file versioning. When rclone uploads a new version of a file it creates a new version of it. Currently rclone only supports retrieving the current version but older versions can be accessed via the Jottacloud Website.</p>
<p>To view your current quota you can use the <code>rclone about remote:</code> command which will display your usage limit (unless it is unlimited) and the current usage.</p>
<p>Note that Jottacloud is case insensitive so you can't have a file called "Hello.doc" and one called "hello.doc".</p>
<p>There are quite a few characters that can't be in Jottacloud file names. Rclone will map these names to and from an identical looking unicode equivalent. For example if a file has a ? in it will be mapped to ？ instead.</p>
<p>Jottacloud only supports filenames up to 255 characters in length.</p>
<h3id="troubleshooting">Troubleshooting</h3>
<p>Jottacloud exhibits some inconsistent behaviours regarding deleted files and folders which may cause Copy, Move and DirMove operations to previously deleted paths to fail. Emptying the trash should help in such cases.</p>
<p>The initial setup for Koofr involves creating an application password for rclone. You can do that by opening the Koofr <ahref="https://app.koofr.net/app/admin/preferences/password">web application</a>, giving the password a nice name like <code>rclone</code> and clicking on generate.</p>
<p>Here is an example of how to make a remote called <code>koofr</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> koofr
Type of storage to configure.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
** See help for koofr backend at: https://rclone.org/koofr/ **
Your Koofr user name
Enter a string value. Press Enter for the default ("").
user> USER@NAME
Your Koofr password for rclone (generate one at https://app.koofr.net/app/admin/preferences/password)
y) Yes type in my own password
g) Generate random password
y/g> y
Enter the password:
password:
Confirm the password:
password:
Edit advanced config? (y/n)
y) Yes
n) No
y/n> n
Remote config
--------------------
[koofr]
type = koofr
baseurl = https://app.koofr.net
user = USER@NAME
password = *** ENCRYPTED ***
--------------------
y) Yes this is OK
e) Edit this remote
d) Delete this remote
y/e/d> y</code></pre>
<p>You can choose to edit advanced config in order to enter your own service URL if you use an on-premise or white label Koofr instance, or choose an alternative mount instead of your primary storage.</p>
<p>Once configured you can then use <code>rclone</code> like this,</p>
<p>List directories in top level of your Koofr</p>
<pre><code>rclone lsd koofr:</code></pre>
<p>List all the files in your Koofr</p>
<pre><code>rclone ls koofr:</code></pre>
<p>To copy a local directory to an Koofr directory called backup</p>
<p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a> the following characters are also replaced:</p>
<p><ahref="https://cloud.mail.ru/">Mail.ru Cloud</a> is a cloud storage provided by a Russian internet company <ahref="https://mail.ru">Mail.Ru Group</a>. The official desktop client is <ahref="https://disk-o.cloud/en">Disk-O:</a>, available on Windows and Mac OS.</p>
<li>Storage keeps hash for all files and performs transparent deduplication, the hash algorithm is a modified SHA1</li>
<li>If a particular file is already present in storage, one can quickly submit file hash instead of long file upload (this optimization is supported by rclone)</li>
<p>Files support a modification time attribute with up to 1 second precision. Directories do not have a modification time, which is shown as "Jan 1 1970".</p>
<p>Hash sums use a custom Mail.ru algorithm based on SHA1. If file size is less than or equal to the SHA1 block size (20 bytes), its hash is simply its data right-padded with zero bytes. Hash sum of a larger file is computed as a SHA1 sum of the file data bytes concatenated with a decimal representation of the data length.</p>
<p>Removing a file or directory actually moves it to the trash, which is not visible to rclone but can be seen in a web browser. The trashed file still occupies part of total quota. If you wish to empty your trash and free some quota, you can use the <code>rclone cleanup remote:</code> command, which will permanently delete all your trashed files. This command does not take any path arguments.</p>
<p>To view your current quota you can use the <code>rclone about remote:</code> command which will display your usage limit (quota) and the current usage.</p>
<p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a> the following characters are also replaced:</p>
<p>File size limits depend on your account. A single file size is limited by 2G for a free account and unlimited for paid tariffs. Please refer to the Mail.ru site for the total uploaded size limits.</p>
<p>Skip full upload if there is another file with same data hash. This feature is called "speedup" or "put by hash". It is especially efficient in case of generally available files like popular books, video or audio clips, because files are searched by hash in all accounts of all mailru users. It is meaningless and ineffective if source file is unique or encrypted. Please note that rclone may need local memory and disk space to calculate content hash in advance and decide whether full upload is required. Also, if rclone does not know file size in advance (e.g. in case of streaming or partial uploads), it will not even try this optimization.</p>
<p>Comma separated list of file name patterns eligible for speedup (put by hash). Patterns are case insensitive and can contain '*' or '?' meta characters.</p>
<p>Comma separated list of internal maintenance flags. This option must not be used by an ordinary user. It is intended only to facilitate remote troubleshooting of backend issues. Strict meaning of flags is not documented and not guaranteed to persist between releases. Quirks will be removed when the backend grows stable. Supported quirks: atomicmkdir binlist unknowndirs</p>
<p><ahref="https://mega.nz/">Mega</a> is a cloud storage and file hosting service known for its security feature where all files are encrypted locally before they are uploaded. This prevents anyone (including employees of Mega) from accessing the files without knowledge of the key used for encryption.</p>
<p>This is an rclone backend for Mega which supports the file transfer features of Mega using the same client side encryption.</p>
<p>Paths are specified as <code>remote:path</code></p>
<p><strong>NOTE:</strong> The encryption keys need to have been already generated after a regular login via the browser, otherwise attempting to use the credentials in <code>rclone</code> will fail.</p>
<p>Mega remotes seem to get blocked (reject logins) under "heavy use". We haven't worked out the exact blocking rules but it seems to be related to fast paced, successive rclone commands.</p>
<p>For example, executing this command 90 times in a row <code>rclone link remote:file</code> will cause the remote to become "blocked". This is not an abnormal situation, for example if you wish to get the public links of a directory with hundred of files... After more or less a week, the remote will remote accept rclone logins normally again.</p>
<p>You can mitigate this issue by mounting the remote it with <code>rclone mount</code>. This will log-in when mounting and a log-out when unmounting only. You can also run <code>rclone rcd</code> and then use <code>rclone rc</code> to run the commands over the API to avoid logging in each time.</p>
<p>Rclone does not currently close mega sessions (you can see them in the web interface), however closing the sessions does not solve the issue.</p>
<p>If you space rclone commands by 3 seconds it will avoid blocking the remote. We haven't identified the exact blocking rules, so perhaps one could execute the command 80 times without waiting and avoid blocking by waiting 3 seconds, then continuing...</p>
<p>Other tools seem not to produce this blocking effect, as they use a different working approach (state-based, using sessionIDs instead of log-in) which isn't compatible with the current stateless rclone approach.</p>
<p>Note that once blocked, the use of other tools (such as megacmd) is not a sure workaround: following megacmd login times have been observed in succession for blocked remote: 7 minutes, 20 min, 30min, 30 min, 30min. Web access looks unaffected though.</p>
<p>Investigation is continuing in relation to workarounds based on timeouts, pacers, retrials and tpslimits - if you discover something relevant, please post on the forum.</p>
<p>So, if rclone was working nicely and suddenly you are unable to log-in and you are sure the user and the password are correct, likely you have got the remote blocked for a while.</p>
<p>Delete files permanently rather than putting them into the trash.</p>
<p>Normally the mega backend will put all deletions into the trash rather than permanently deleting them. If you specify this then rclone will permanently delete objects instead.</p>
<p>This backend uses the <ahref="https://github.com/t3rm1n4l/go-mega">go-mega go library</a> which is an opensource go library implementing the Mega API. There doesn't appear to be any documentation for the mega protocol beyond the <ahref="https://github.com/meganz/sdk">mega C++ SDK</a> source code so there are likely quite a few errors still remaining in this library.</p>
<p>The memory backend behaves like a bucket based remote (e.g. like s3). Because it has no parameters you can just use it with the <code>:memory:</code> remote name.</p>
<p>Paths are specified as <code>remote:container</code> (or <code>remote:</code> for the <code>lsd</code> command.) You may put subdirectories in too, e.g. <code>remote:container/path/to/dir</code>.</p>
<p>This remote supports <code>--fast-list</code> which allows you to use fewer transactions in exchange for more memory. See the <ahref="https://rclone.org/docs/#fast-list">rclone docs</a> for more details.</p>
<p>The modified time is stored as metadata on the object with the <code>mtime</code> key. It is stored using RFC3339 Format time with nanosecond precision. The metadata is supplied during directory listings so there is no overhead to using it.</p>
<p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a> the following characters are also replaced:</p>
<p>MD5 hashes are stored with blobs. However blobs that were uploaded in chunks only have an MD5 if the source remote was capable of MD5 hashes, e.g. the local disk.</p>
<h3id="authenticating-with-azure-blob-storage">Authenticating with Azure Blob Storage</h3>
<p>Rclone has 3 ways of authenticating with Azure Blob Storage:</p>
<h4id="account-and-key">Account and Key</h4>
<p>This is the most straight forward and least flexible way. Just fill in the <code>account</code> and <code>key</code> lines and leave the rest blank.</p>
<p>An account level SAS URL or container level SAS URL can be obtained from the Azure portal or the Azure Storage Explorer. To get a container level SAS URL right click on a container in the Azure Blob explorer in the Azure portal.</p>
<pre><code>rclone ls azureblob:othercontainer</code></pre>
<p>Container level SAS URLs are useful for temporarily allowing third parties access to a single container or putting credentials into an untrusted environment such as a CI build server.</p>
<p>See <ahref="https://docs.microsoft.com/en-us/azure/storage/common/storage-auth-aad-rbac-cli">Use Azure CLI to assign an Azure role for access to blob and queue data</a> for more details.</p>
<p>Use a managed service identity to authenticate (only works in Azure)</p>
<p>When true, use a <ahref="https://docs.microsoft.com/en-us/azure/active-directory/managed-identities-azure-resources/">managed service identity</a> to authenticate to Azure Storage instead of a SAS token or account key.</p>
<p>If the VM(SS) on which this program is running has a system-assigned identity, it will be used by default. If the resource has no system-assigned but exactly one user-assigned identity, the user-assigned identity will be used by default. If the resource has multiple user-assigned identities, the identity to use must be explicitly specified using exactly one of the msi_object_id, msi_client_id, or msi_mi_res_id parameters.</p>
<p>This sets the number of blobs requested in each listing chunk. Default is the maximum, 5000. "List blobs" requests are permitted 2 minutes per megabyte to complete. If an operation is taking longer than 2 minutes per megabyte on average, it will time out ( <ahref="https://docs.microsoft.com/en-us/rest/api/storageservices/setting-timeouts-for-blob-service-operations#exceptions-to-default-timeout-interval">source</a> ). This can be used to limit the number of blobs items to return, to avoid the time out.</p>
<p>Archived blobs can be restored by setting access tier to hot or cool. Leave blank if you intend to use default access tier, which is set at account level</p>
<p>If there is no "access tier" specified, rclone doesn't apply any tier. rclone performs "Set Tier" operation on blobs while uploading, if objects are not modified, specifying "access tier" to new one will have no effect. If blobs are in "archive tier" at remote, trying to perform data transfer operations from remote will not be allowed. User should first restore by tiering blob to "Hot" or "Cool".</p>
<p>Delete archive tier blobs before overwriting.</p>
<p>Archive tier blobs cannot be updated. So without this flag, if you attempt to update an archive tier blob, then rclone will produce the error:</p>
<pre><code>can't update archive tier blob without --azureblob-archive-tier-delete</code></pre>
<p>With this flag set then before rclone attempts to overwrite an archive tier blob, it will delete the existing blob before uploading its replacement. This has the potential for data loss if the upload fails (unlike updating a normal blob) and also may cost more since deleting archive tier blobs early may be chargable.</p>
<p>Normally rclone will calculate the MD5 checksum of the input before uploading it so it can add it to metadata on the object. This is great for data integrity checking but can cause long delays for large files to start uploading.</p>
<p>How often internal memory buffer pools will be flushed. Uploads which requires additional buffers (f.e multipart) will use memory pool for allocations. This option controls how often unused buffers will be removed from the pool.</p>
<p><code>rclone about</code> is not supported by the Microsoft Azure Blob storage backend. Backends without this capability cannot determine free space for an rclone mount or use policy <code>mfs</code> (most free space) as a member of an rclone union remote.</p>
<p>See <ahref="https://rclone.org/overview/#optional-features">List of backends that do not support rclone about</a> See <ahref="https://rclone.org/commands/rclone_about/">rclone about</a></p>
<p>You can test rclone with storage emulator locally, to do this make sure azure storage emulator installed locally and set up a new remote with <code>rclone config</code> follow instructions described in introduction, set <code>use_emulator</code> config as <code>true</code>, you do not need to provide default account name or key if using emulator.</p>
<p>The initial setup for OneDrive involves getting a token from Microsoft which you need to do in your browser. <code>rclone config</code> walks you through it.</p>
<p>Note that rclone runs a webserver on your local machine to collect the token as returned from Microsoft. This only runs from the moment it opens your browser to the moment you get back the verification code. This is on <code>http://127.0.0.1:53682/</code> and this it may require you to unblock it temporarily if you are running a host firewall.</p>
<p>You can use your own Client ID if the default (<code>client_id</code> left blank) one doesn't work for you or you see lots of throttling. The default Client ID and Key is shared by all rclone users when performing requests.</p>
<li>Enter a name for your app, choose account type <code>Accounts in any organizational directory (Any Azure AD directory - Multitenant) and personal Microsoft accounts (e.g. Skype, Xbox)</code>, select <code>Web</code> in <code>Redirect URI</code>, then type (do not copy and paste) <code>http://localhost:53682/</code> and click Register. Copy and keep the <code>Application (client) ID</code> under the app name for later use.</li>
<li>Under <code>manage</code> select <code>Certificates & secrets</code>, click <code>New client secret</code>. Copy and keep that secret for later use.</li>
<li>Under <code>manage</code> select <code>API permissions</code>, click <code>Add a permission</code> and select <code>Microsoft Graph</code> then select <code>delegated permissions</code>.</li>
<li>Search and select the following permissions: <code>Files.Read</code>, <code>Files.ReadWrite</code>, <code>Files.Read.All</code>, <code>Files.ReadWrite.All</code>, <code>offline_access</code>, <code>User.Read</code>. Once selected click <code>Add permissions</code> at the bottom.</li>
<p>Now the application is complete. Run <code>rclone config</code> to create or edit a OneDrive remote. Supply the app ID and password as Client ID and Secret, respectively. rclone will walk you through the remaining steps.</p>
<p>OneDrive personal supports SHA1 type hashes. OneDrive for business and Sharepoint Server support <ahref="https://docs.microsoft.com/en-us/onedrive/developer/code-snippets/quickxorhash">QuickXorHash</a>.</p>
<p>For all types of OneDrive you can use the <code>--checksum</code> flag.</p>
<p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a> the following characters are also replaced:</p>
<p>Any files you delete with rclone will end up in the trash. Microsoft doesn't provide an API to permanently delete files, nor to empty the trash, so you will have to do that with one of Microsoft's apps or via the OneDrive website.</p>
<p>Above this size files will be chunked - must be multiple of 320k (327,680 bytes) and should not exceed 250M (262,144,000 bytes) else you may encounter "Microsoft.SharePoint.Client.InvalidClientQueryException: The request message is too big." Note that the chunks will be buffered into memory.</p>
<p>By default rclone will hide OneNote files in directory listings because operations like "Open" and "Update" won't work on them. But this behaviour may also prevent you from deleting them. If you want to delete OneNote files or otherwise want them to show up in directory listing, set this option.</p>
<p>Allow server-side operations (e.g. copy) to work across different onedrive configs.</p>
<p>This will only work if you are copying between two OneDrive <em>Personal</em> drives AND the files to copy are already shared between them. In other cases, rclone will fall back to normal copy (which will be slightly slower).</p>
<p>Set the scope of the links created by the link command.</p>
<ul>
<li>Config: link_scope</li>
<li>Env Var: RCLONE_ONEDRIVE_LINK_SCOPE</li>
<li>Type: string</li>
<li>Default: "anonymous"</li>
<li>Examples:
<ul>
<li>"anonymous"
<ul>
<li>Anyone with the link has access, without needing to sign in. This may include people outside of your organization. Anonymous link support may be disabled by an administrator.</li>
</ul></li>
<li>"organization"
<ul>
<li>Anyone signed into your organization (tenant) can use the link to get access. Only available in OneDrive for Business and SharePoint.</li>
<p>If you don't use rclone for 90 days the refresh token will expire. This will result in authorization problems. This is easy to fix by running the <code>rclone config reconnect remote:</code> command to get a new token and refresh token.</p>
<p>Note that OneDrive is case insensitive so you can't have a file called "Hello.doc" and one called "hello.doc".</p>
<p>There are quite a few characters that can't be in OneDrive file names. These can't occur on Windows platforms, but on non-Windows platforms they are common. Rclone will map these names to and from an identical looking unicode equivalent. For example if a file has a <code>?</code> in it will be mapped to <code>？</code> instead.</p>
<p>The largest allowed file size is 250GB for both OneDrive Personal and OneDrive for Business <ahref="https://support.microsoft.com/en-us/office/invalid-file-names-and-file-types-in-onedrive-and-sharepoint-64883a5d-228e-48f5-b3d2-eb39e07630fa?ui=en-us&rs=en-us&ad=us#individualfilesize">(Updated 13 Jan 2021)</a>.</p>
<p>The entire path, including the file name, must contain fewer than 400 characters for OneDrive, OneDrive for Business and SharePoint Online. If you are encrypting file and folder names with rclone, you may want to pay attention to this limitation because the encrypted names are typically longer than the original ones.</p>
<p>OneDrive seems to be OK with at least 50,000 files in a folder, but at 100,000 rclone will get errors listing the directory like <code>couldn’t list files: UnknownError:</code>. See <ahref="https://github.com/rclone/rclone/issues/2707">#2707</a> for more info.</p>
<p>An official document about the limitations for different types of OneDrive can be found <ahref="https://support.office.com/en-us/article/invalid-file-names-and-file-types-in-onedrive-onedrive-for-business-and-sharepoint-64883a5d-228e-48f5-b3d2-eb39e07630fa">here</a>.</p>
<p>Every change in a file OneDrive causes the service to create a new version of the the file. This counts against a users quota. For example changing the modification time of a file creates a second version, so the file apparently uses twice the space.</p>
<p>For example the <code>copy</code> command is affected by this as rclone copies the file and then afterwards sets the modification time to match the source file which uses another version.</p>
<p>You can use the <code>rclone cleanup</code> command (see below) to remove all old versions.</p>
<p>Or you can set the <code>no_versions</code> parameter to <code>true</code> and rclone will remove versions after operations which create new versions. This takes extra transactions so only enable it if you need it.</p>
<p><strong>Note</strong> At the time of writing Onedrive Personal creates versions (but not for setting the modification time) but the API for removing them returns "API not found" so cleanup and <code>no_versions</code> should not be used on Onedrive Personal.</p>
<p>Starting October 2018, users will no longer be able to disable versioning by default. This is because Microsoft has brought an <ahref="https://techcommunity.microsoft.com/t5/Microsoft-OneDrive-Blog/New-Updates-to-OneDrive-and-SharePoint-Team-Site-Versioning/ba-p/204390">update</a> to the mechanism. To change this new default setting, a PowerShell command is required to be run by a SharePoint admin. If you are an admin, you can run these commands in PowerShell to change that setting:</p>
<li><code>Connect-SPOService -Url https://YOURSITE-admin.sharepoint.com -Credential YOU@YOURSITE.COM</code> (replacing <code>YOURSITE</code>, <code>YOU</code>, <code>YOURSITE.COM</code> with the actual values; this will prompt for your credentials)</li>
<p><em>Below are the steps for normal users to disable versioning. If you don't see the "No Versioning" option, make sure the above requirements are met.</em></p>
<li>Under Document Version History select the option No versioning. Note: This will disable the creation of new file versions, but will not remove any previous versions. Your documents are safe.</li>
<p>OneDrive supports <code>rclone cleanup</code> which causes rclone to look through every file under the path supplied and delete all version but the current version. Because this involves traversing all the files, then querying each file for versions it can be quite slow. Rclone does <code>--checkers</code> tests in parallel. The command also supports <code>-i</code> which is a great way to see what it would do.</p>
<pre><code>rclone cleanup -i remote:path/subdir # interactively remove all old version for path/subdir
rclone cleanup remote:path/subdir # unconditionally remove all old version for path/subdir</code></pre>
<p><strong>NB</strong> Onedrive personal can't currently delete versions</p>
<p>It is a <ahref="https://github.com/OneDrive/onedrive-api-docs/issues/935#issuecomment-441741631">known</a> issue that Sharepoint (not OneDrive or OneDrive for Business) silently modifies uploaded files, mainly Office files (.docx, .xlsx, etc.), causing file size and hash checks to fail. There are also other situations that will cause OneDrive to report inconsistent file sizes. To use rclone with such affected files on Sharepoint, you may disable these checks with the following command line arguments:</p>
<p>Alternatively, if you have write access to the OneDrive files, it may be possible to fix this problem for certain files, by attempting the steps below. Open the web interface for <ahref="https://onedrive.live.com">OneDrive</a> and find the affected files (which will be in the error messages/log for rclone). Simply click on each of these files, causing OneDrive to open them on the web. This will cause each file to be converted in place to a format that is functionally equivalent but which will no longer trigger the size discrepancy. Once all problematic files are converted you will no longer need the ignore options above.</p>
<h4id="replacingdeleting-existing-files-on-sharepoint-gets-item-not-found">Replacing/deleting existing files on Sharepoint gets "item not found"</h4>
<p>It is a <ahref="https://github.com/OneDrive/onedrive-api-docs/issues/1068">known</a> issue that Sharepoint (not OneDrive or OneDrive for Business) may return "item not found" errors when users try to replace or delete uploaded files; this seems to mainly affect Office files (.docx, .xlsx, etc.). As a workaround, you may use the <code>--backup-dir <BACKUP_DIR></code> command line argument so rclone moves the files to be replaced/deleted into a given backup directory (instead of directly replacing/deleting them). For example, to instruct rclone to move the files into the directory <code>rclone-backup-dir</code> on backend <code>mysharepoint</code>, you may use:</p>
Description: Using application 'rclone' is currently not supported for your organization [YOUR_ORGANIZATION] because it is in an unmanaged state. An administrator needs to claim ownership of the company by DNS validation of [YOUR_ORGANIZATION] before the application rclone can be provisioned.</code></pre>
<p>This means that rclone can't use the OneDrive for Business API with your account. You can't do much about it, maybe write an email to your admins.</p>
Description: Due to a configuration change made by your administrator, or because you moved to a new location, you must use multi-factor authentication to access '...'.</code></pre>
<p>If you see the error above after enabling multi-factor authentication for your account, you can fix it by refreshing your OAuth refresh token. To do that, run <code>rclone config</code>, and choose to edit your OneDrive backend. Then, you don't need to actually make any changes until you reach this question: <code>Already have a token - refresh?</code>. For this question, answer <code>y</code> and go through the process to refresh your token, just like the first time the backend is configured. After this, rclone should work again for this backend.</p>
<p>OpenDrive allows modification times to be set on objects accurate to 1 second. These will be used to detect whether objects need syncing or not.</p>
<p>File names can also not begin or end with the following characters. These only get replaced if they are the first or last character in the name:</p>
<p>Note that OpenDrive is case insensitive so you can't have a file called "Hello.doc" and one called "hello.doc".</p>
<p>There are quite a few characters that can't be in OpenDrive file names. These can't occur on Windows platforms, but on non-Windows platforms they are common. Rclone will map these names to and from an identical looking unicode equivalent. For example if a file has a <code>?</code> in it will be mapped to <code>？</code> instead.</p>
<p><code>rclone about</code> is not supported by the OpenDrive backend. Backends without this capability cannot determine free space for an rclone mount or use policy <code>mfs</code> (most free space) as a member of an rclone union remote.</p>
<p>See <ahref="https://rclone.org/overview/#optional-features">List of backends that do not support rclone about</a> See <ahref="https://rclone.org/commands/rclone_about/">rclone about</a></p>
<p>Paths are specified as <code>remote:bucket</code> (or <code>remote:</code> for the <code>lsd</code> command.) You may put subdirectories in too, e.g. <code>remote:bucket/path/to/dir</code>.</p>
<p>This remote supports <code>--fast-list</code> which allows you to use fewer transactions in exchange for more memory. See the <ahref="https://rclone.org/docs/#fast-list">rclone docs</a> for more details.</p>
<p>rclone supports multipart uploads with QingStor which means that it can upload files bigger than 5GB. Note that files uploaded with multipart upload don't have an MD5SUM.</p>
<p>Note that incomplete multipart uploads older than 24 hours can be removed with <code>rclone cleanup remote:bucket</code> just for one bucket <code>rclone cleanup remote:</code> for all buckets. QingStor does not ever remove incomplete multipart uploads so it may be necessary to run this from time to time.</p>
<p>With QingStor you can list buckets (<code>rclone lsd</code>) using any zone, but you can only access the content of a bucket from the zone it was created in. If you attempt to access a bucket from the wrong zone, you will get an error, <code>incorrect zone, the bucket is not in 'XXX' zone</code>.</p>
<p>The control characters 0x00-0x1F and / are replaced as in the <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a>. Note that 0x7F is not replaced.</p>
<p>NB if you set this to > 1 then the checksums of multipart uploads become corrupted (the uploads themselves are not corrupted though).</p>
<p>If you are uploading small numbers of large files over high-speed links and these uploads do not fully utilize your bandwidth, then increasing this may help to speed up the transfers.</p>
<p><code>rclone about</code> is not supported by the qingstor backend. Backends without this capability cannot determine free space for an rclone mount or use policy <code>mfs</code> (most free space) as a member of an rclone union remote.</p>
<p>See <ahref="https://rclone.org/overview/#optional-features">List of backends that do not support rclone about</a> See <ahref="https://rclone.org/commands/rclone_about/">rclone about</a></p>
<p>Paths are specified as <code>remote:container</code> (or <code>remote:</code> for the <code>lsd</code> command.) You may put subdirectories in too, e.g. <code>remote:container/path/to/dir</code>.</p>
echo "Please enter your OpenStack Password: "
read -sr OS_PASSWORD_INPUT
export OS_PASSWORD=$OS_PASSWORD_INPUT
export OS_REGION_NAME="SBG1"
if [ -z "$OS_REGION_NAME" ]; then unset OS_REGION_NAME; fi</code></pre>
<p>The config file needs to look something like this where <code>$OS_USERNAME</code> represents the value of the <code>OS_USERNAME</code> variable - <code>123abc567xy</code> in the example above.</p>
<pre><code>[remote]
type = swift
user = $OS_USERNAME
key = $OS_PASSWORD
auth = $OS_AUTH_URL
tenant = $OS_TENANT_NAME</code></pre>
<p>Note that you may (or may not) need to set <code>region</code> too - try without first.</p>
<h3id="configuration-from-the-environment">Configuration from the environment</h3>
<p>If you prefer you can configure rclone to use swift using a standard set of OpenStack environment variables.</p>
<p>When you run through the config, make sure you choose <code>true</code> for <code>env_auth</code> and leave everything else blank.</p>
<p>rclone will then set any empty config parameters from the environment using standard OpenStack environment variables. There is <ahref="https://godoc.org/github.com/ncw/swift#Connection.ApplyEnvironment">a list of the variables</a> in the docs for the swift library.</p>
<h3id="using-an-alternate-authentication-method">Using an alternate authentication method</h3>
<p>If your OpenStack installation uses a non-standard authentication method that might not be yet supported by rclone or the underlying swift library, you can authenticate externally (e.g. calling manually the <code>openstack</code> commands to get a token). Then, you just need to pass the two configuration variables <code>auth_token</code> and <code>storage_url</code>. If they are both provided, the other variables are ignored. rclone will not try to authenticate but instead assume it is already authenticated and use these two variables to access the OpenStack installation.</p>
<p>This remote supports <code>--fast-list</code> which allows you to use fewer transactions in exchange for more memory. See the <ahref="https://rclone.org/docs/#fast-list">rclone docs</a> for more details.</p>
<p>As noted below, the modified time is stored on metadata on the object. It is used by default for all operations that require checking the time a file was last updated. It allows rclone to treat the remote more like a true filesystem, but it is inefficient because it requires an extra API call to retrieve the metadata.</p>
<p>For many operations, the time the object was last uploaded to the remote is sufficient to determine if it is "dirty". By using <code>--update</code> along with <code>--use-server-modtime</code>, you can avoid the extra API call and simply upload files whose local modtime is newer than the time it was last uploaded.</p>
<p>The storage policy to use when creating a new container</p>
<p>This applies the specified storage policy when creating a new container. The policy cannot be changed afterwards. The allowed configuration values and their meaning depend on your Swift storage provider.</p>
<p>The Swift API doesn't return a correct MD5SUM for segmented files (Dynamic or Static Large Objects) so rclone won't check or use the MD5SUM for these.</p>
<h4id="rclone-gives-failed-to-create-file-system-for-remote-bad-request">Rclone gives Failed to create file system for "remote:": Bad Request</h4>
<p>Due to an oddity of the underlying swift library, it gives a "Bad Request" error rather than a more sensible error when the authentication fails for Swift.</p>
<h4id="rclone-gives-failed-to-create-file-system-response-didnt-have-storage-url-and-auth-token">Rclone gives Failed to create file system: Response didn't have storage url and auth token</h4>
<p>The initial setup for pCloud involves getting a token from pCloud which you need to do in your browser. <code>rclone config</code> walks you through it.</p>
<p>Here is an example of how to make a remote called <code>remote</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> remote
Type of storage to configure.
Choose a number from below, or type in your own value
<p>See the <ahref="https://rclone.org/remote_setup/">remote setup docs</a> for how to set it up on a machine with no Internet browser available.</p>
<p>Note that rclone runs a webserver on your local machine to collect the token as returned from pCloud. This only runs from the moment it opens your browser to the moment you get back the verification code. This is on <code>http://127.0.0.1:53682/</code> and this it may require you to unblock it temporarily if you are running a host firewall.</p>
<p>Once configured you can then use <code>rclone</code> like this,</p>
<p>List directories in top level of your pCloud</p>
<pre><code>rclone lsd remote:</code></pre>
<p>List all the files in your pCloud</p>
<pre><code>rclone ls remote:</code></pre>
<p>To copy a local directory to an pCloud directory called backup</p>
<p>pCloud allows modification times to be set on objects accurate to 1 second. These will be used to detect whether objects need syncing or not. In order to set a Modification time pCloud requires the object be re-uploaded.</p>
<p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a> the following characters are also replaced:</p>
<p>Deleted files will be moved to the trash. Your subscription level will determine how long items stay in the trash. <code>rclone cleanup</code> can be used to empty the trash.</p>
<p>You can set the <code>root_folder_id</code> for rclone. This is the directory (identified by its <code>Folder ID</code>) that rclone considers to be the root of your pCloud drive.</p>
<p>Normally you will leave this blank and rclone will determine the correct root to use itself.</p>
<p>However you can set this to restrict rclone to a specific folder hierarchy.</p>
<p>In order to do this you will have to find the <code>Folder ID</code> of the directory you wish rclone to display. This will be the <code>folder</code> field of the URL when you open the relevant folder in the pCloud web interface.</p>
<p>So if the folder you want rclone to use has a URL which looks like <code>https://my.pcloud.com/#page=filemanager&folder=5xxxxxxxx8&tpl=foldergrid</code> in the browser, then you use <code>5xxxxxxxx8</code> as the <code>root_folder_id</code> in the config.</p>
<p>This is normally set when rclone initially does the oauth connection, however you will need to set it by hand if you are using remote config with rclone authorize.</p>
<p>The initial setup for <ahref="https://premiumize.me/">premiumize.me</a> involves getting a token from premiumize.me which you need to do in your browser. <code>rclone config</code> walks you through it.</p>
<p>Here is an example of how to make a remote called <code>remote</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> remote
Type of storage to configure.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
[snip]
XX / premiumize.me
\ "premiumizeme"
[snip]
Storage> premiumizeme
** See help for premiumizeme backend at: https://rclone.org/premiumizeme/ **
Remote config
Use auto config?
* Say Y if not sure
* Say N if you are working on a remote or headless machine
y) Yes
n) No
y/n> y
If your browser doesn't open automatically go to the following link: http://127.0.0.1:53682/auth
<p>See the <ahref="https://rclone.org/remote_setup/">remote setup docs</a> for how to set it up on a machine with no Internet browser available.</p>
<p>Note that rclone runs a webserver on your local machine to collect the token as returned from premiumize.me. This only runs from the moment it opens your browser to the moment you get back the verification code. This is on <code>http://127.0.0.1:53682/</code> and this it may require you to unblock it temporarily if you are running a host firewall.</p>
<p>Once configured you can then use <code>rclone</code> like this,</p>
<p>List directories in top level of your premiumize.me</p>
<pre><code>rclone lsd remote:</code></pre>
<p>List all the files in your premiumize.me</p>
<pre><code>rclone ls remote:</code></pre>
<p>To copy a local directory to an premiumize.me directory called backup</p>
<p>premiumize.me does not support modification times or hashes, therefore syncing will default to <code>--size-only</code> checking. Note that using <code>--update</code> will work.</p>
<p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a> the following characters are also replaced:</p>
<p>Note that premiumize.me is case insensitive so you can't have a file called "Hello.doc" and one called "hello.doc".</p>
<p>premiumize.me file names can't have the <code>\</code> or <code>"</code> characters in. rclone maps these to and from an identical looking unicode equivalents <code>＼</code> and <code>＂</code></p>
<p>The initial setup for put.io involves getting a token from put.io which you need to do in your browser. <code>rclone config</code> walks you through it.</p>
<p>Here is an example of how to make a remote called <code>remote</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> putio
Type of storage to configure.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
[snip]
XX / Put.io
\ "putio"
[snip]
Storage> putio
** See help for putio backend at: https://rclone.org/putio/ **
Remote config
Use auto config?
* Say Y if not sure
* Say N if you are working on a remote or headless machine
y) Yes
n) No
y/n> y
If your browser doesn't open automatically go to the following link: http://127.0.0.1:53682/auth
<p>Note that rclone runs a webserver on your local machine to collect the token as returned from Google if you use auto config mode. This only runs from the moment it opens your browser to the moment you get back the verification code. This is on <code>http://127.0.0.1:53682/</code> and this it may require you to unblock it temporarily if you are running a host firewall, or use manual mode.</p>
<p>You can then use it like this,</p>
<p>List directories in top level of your put.io</p>
<p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a> the following characters are also replaced:</p>
<p>This is a backend for the <ahref="https://www.seafile.com/">Seafile</a> storage service: - It works with both the free community edition or the professional edition. - Seafile versions 6.x and 7.x are all supported. - Encrypted libraries are also supported. - It supports 2FA enabled users</p>
<h3id="root-mode-vs-library-mode">Root mode vs Library mode</h3>
<p>There are two distinct modes you can setup your remote: - you point your remote to the <strong>root of the server</strong>, meaning you don't specify a library during the configuration: Paths are specified as <code>remote:library</code>. You may put subdirectories in too, e.g. <code>remote:library/path/to/dir</code>. - you point your remote to a specific library during the configuration: Paths are specified as <code>remote:path/to/dir</code>. <strong>This is the recommended mode when using encrypted libraries</strong>. (<em>This mode is possibly slightly faster than the root mode</em>)</p>
<p>This will guide you through an interactive setup process. To authenticate you will need the URL of your server, your email (or username) and your password.</p>
<p>Here's an example of a configuration in library mode with a user that has the two-factor authentication enabled. Your 2FA code will be asked at the end of the configuration, and will attempt to authenticate you:</p>
<p>You specified <code>My Library</code> during the configuration. The root of the remote is pointing at the root of the library <code>My Library</code>:</p>
<p>Seafile version 7+ supports <code>--fast-list</code> which allows you to use fewer transactions in exchange for more memory. See the <ahref="https://rclone.org/docs/#fast-list">rclone docs</a> for more details. Please note this is not supported on seafile server version 6.x</p>
<p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default restricted characters set</a> the following characters are also replaced:</p>
<p>Please note a share link is unique for each file or directory. If you run a link command on a file/dir that has already been shared, you will get the exact same link.</p>
<h3id="compatibility">Compatibility</h3>
<p>It has been actively tested using the <ahref="https://github.com/haiwen/seafile-docker">seafile docker image</a> of these versions: - 6.3.4 community edition - 7.0.5 community edition - 7.1.3 community edition</p>
<p>Paths are specified as <code>remote:path</code>. If the path does not begin with a <code>/</code> it is relative to the home directory of the user. An empty path <code>remote:</code> refers to the user's home directory.</p>
<p>"Note that some SFTP servers will need the leading / - Synology is a good example of this. rsync.net, on the other hand, requires users to OMIT the leading /.</p>
<p>Here is an example of making an SFTP configuration. First run</p>
<pre><code>rclone config</code></pre>
<p>This will guide you through an interactive setup process.</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> remote
Type of storage to configure.
Choose a number from below, or type in your own value
[snip]
XX / SSH/SFTP Connection
\ "sftp"
[snip]
Storage> sftp
SSH host to connect to
Choose a number from below, or type in your own value
<p>Key files should be PEM-encoded private key files. For instance <code>/home/$USER/.ssh/id_rsa</code>. Only unencrypted OpenSSH or PEM encrypted files are supported.</p>
<p>The key file can be specified in either an external file (key_file) or contained within the rclone config file (key_pem). If using key_pem in the config file, the entry should be on a single line with new line ('' or '') separating lines. i.e.</p>
<p>You can also specify <code>key_use_agent</code> to force the usage of an ssh-agent. In this case <code>key_file</code> or <code>key_pem</code> can also be specified to force the usage of a specific key in the ssh-agent.</p>
<p>By default rclone will not check the server's host key for validation. This can allow an attacker to replace a server with their own and if you use password authentication then this can lead to that password being exposed.</p>
<p>Host key matching, using standard <code>known_hosts</code> files can be turned on by enabling the <code>known_hosts_file</code> option. This can point to the file maintained by <code>OpenSSH</code> or can point to a unique file.</p>
<li><code>rclone</code> will not <em>manage</em> this file for you. If the key is missing or wrong then the connection will be refused.</li>
<li>If the server is set up for a certificate host key then the entry in the <code>known_hosts</code> file <em>must</em> be the <code>@cert-authority</code> entry for the CA</li>
<li>Unlike <code>OpenSSH</code>, the libraries used by <code>rclone</code> do not permit (at time of writing) multiple host keys to be listed for a server. Only the first entry is used.</li>
</ul>
<p>If the host key provided by the server does not match the one in the file (or is missing) then the connection will be aborted and an error returned such as</p>
<pre><code>NewFs: couldn't connect SSH: ssh: handshake failed: ssh: no authorities for hostname: example.com:22</code></pre>
<p>then it is likely the server has presented a CA signed host certificate and you will need to add the appropriate <code>@cert-authority</code> entry.</p>
<p>The <code>known_hosts_file</code> setting can be set during <code>rclone config</code> as an advanced option.</p>
<p>Note that there seem to be various problems with using an ssh-agent on macOS due to recent changes in the OS. The most effective work-around seems to be to start an ssh-agent in each session, e.g.</p>
<p>Some SFTP servers disable setting/modifying the file modification time after upload (for example, certain configurations of ProFTPd with mod_sftp). If you are using one of these servers, you can set the option <code>set_modtime = false</code> in your RClone backend configuration to disable this behaviour.</p>
<p>When key-file is also set, the ".pub" file of the specified key-file is read and only the associated key is requested from the ssh-agent. This allows to avoid <code>Too many authentication failures for *username*</code> errors when the ssh-agent contains many keys.</p>
<p>Disable the execution of SSH commands to determine if remote file hashing is available. Leave blank or set to false to enable hashing (recommended), set to true to disable hashing.</p>
<p>Some servers limit the amount of open files and calling Stat after opening the file will throw an error from the server. Setting this flag will call Fstat instead of Stat which is called on an already open file handle.</p>
<p>It has been found that this helps with IBM Sterling SFTP servers which have "extractability" level set to 1 which means only 1 file can be opened at any given time.</p>
<p>Normally concurrent reads are safe to use and not using them will degrade performance, so this option is disabled by default.</p>
<p>Some servers limit the amount number of times a file can be downloaded. Using concurrent reads can trigger this limit, so if you have a server which returns</p>
<pre><code>Failed to copy: file does not exist</code></pre>
<p>Then you may need to enable this flag.</p>
<p>If concurrent reads are disabled, the use_fstat option is ignored.</p>
<p>SFTP supports checksums if the same login has shell access and <code>md5sum</code> or <code>sha1sum</code> as well as <code>echo</code> are in the remote's PATH. This remote checksumming (file hashing) is recommended and enabled by default. Disabling the checksumming may be required if you are connecting to SFTP servers which are not under your control, and to which the execution of remote commands is prohibited. Set the configuration option <code>disable_hashcheck</code> to <code>true</code> to disable checksumming.</p>
<p>SFTP also supports <code>about</code> if the same login has shell access and <code>df</code> are in the remote's PATH. <code>about</code> will return the total space, free space, and used space on the remote for the disk of the specified path on the remote or, if not set, the disk of the root on the remote. <code>about</code> will fail if it does not have shell access or if <code>df</code> is not in the remote's PATH.</p>
<p>Note that some SFTP servers (e.g. Synology) the paths are different for SSH and SFTP so the hashes can't be calculated properly. For them using <code>disable_hashcheck</code> is a good idea.</p>
<p>The Go SSH library disables the use of the aes128-cbc cipher by default, due to security concerns. This can be re-enabled on a per-connection basis by setting the <code>use_insecure_cipher</code> setting in the configuration file to <code>true</code>. Further details on the insecurity of this cipher can be found <ahref="http://www.isg.rhul.ac.uk/~kp/SandPfinal.pdf">in this paper</a>.</p>
<p>SFTP isn't supported under plan9 until <ahref="https://github.com/pkg/sftp/issues/156">this issue</a> is fixed.</p>
<p>Note that since SFTP isn't HTTP based the following flags don't work with it: <code>--dump-headers</code>, <code>--dump-bodies</code>, <code>--dump-auth</code></p>
<p>Note that <code>--timeout</code> isn't supported (but <code>--contimeout</code> is).</p>
<p><ahref="https://sugarsync.com">SugarSync</a> is a cloud service that enables active synchronization of files across computers and other devices for file backup, access, syncing, and sharing.</p>
<p>The initial setup for SugarSync involves getting a token from SugarSync which you can do with rclone. <code>rclone config</code> walks you through it.</p>
<p>Here is an example of how to make a remote called <code>remote</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> remote
Type of storage to configure.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
[snip]
XX / Sugarsync
\ "sugarsync"
[snip]
Storage> sugarsync
** See help for sugarsync backend at: https://rclone.org/sugarsync/ **
Sugarsync App ID.
Leave blank to use rclone's.
Enter a string value. Press Enter for the default ("").
app_id>
Sugarsync Access Key ID.
Leave blank to use rclone's.
Enter a string value. Press Enter for the default ("").
access_key_id>
Sugarsync Private Access Key
Leave blank to use rclone's.
Enter a string value. Press Enter for the default ("").
private_access_key>
Permanently delete files if true
otherwise put them in the deleted files.
Enter a boolean value (true or false). Press Enter for the default ("false").
hard_delete>
Edit advanced config? (y/n)
y) Yes
n) No (default)
y/n> n
Remote config
Username (email address)> nick@craig-wood.com
Your Sugarsync password is only required during setup and will not be stored.
<p><strong>NB</strong> you can't create files in the top level folder you have to create a folder, which rclone will create as a "Sync Folder" with SugarSync.</p>
<p>SugarSync does not support modification times or hashes, therefore syncing will default to <code>--size-only</code> checking. Note that using <code>--update</code> will work as rclone can read the time files were uploaded.</p>
<p>However you can supply the flag <code>--sugarsync-hard-delete</code> or set the config parameter <code>hard_delete = true</code> if you would like files to be deleted straight away.</p>
<p><code>rclone about</code> is not supported by the SugarSync backend. Backends without this capability cannot determine free space for an rclone mount or use policy <code>mfs</code> (most free space) as a member of an rclone union remote.</p>
<p>See <ahref="https://rclone.org/overview/#optional-features">List of backends that do not support rclone about</a> See <ahref="https://rclone.org/commands/rclone_about/">rclone about</a></p>
<p><ahref="https://tardigrade.io">Tardigrade</a> is an encrypted, secure, and cost-effective object storage service that enables you to store, back up, and archive large amounts of data in a decentralized manner.</p>
<p>To make a new Tardigrade configuration you need one of the following: * Access Grant that someone else shared with you. * <ahref="https://documentation.tardigrade.io/getting-started/uploading-your-first-object/create-an-api-key">API Key</a> of a Tardigrade project you are a member of.</p>
<p>Here is an example of how to make a remote called <code>remote</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<h3id="setup-with-access-grant">Setup with access grant</h3>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> remote
Type of storage to configure.
Enter a string value. Press Enter for the default ("").
Choose a number from below, or type in your own value
[snip]
XX / Tardigrade Decentralized Cloud Storage
\ "tardigrade"
[snip]
Storage> tardigrade
** See help for tardigrade backend at: https://rclone.org/tardigrade/ **
Choose an authentication method.
Enter a string value. Press Enter for the default ("existing").
Choose a number from below, or type in your own value
1 / Use an existing access grant.
\ "existing"
2 / Create a new access grant from satellite address, API key, and passphrase.
\ "new"
provider> existing
Access Grant.
Enter a string value. Press Enter for the default ("").
<p>Paths are specified as <code>remote:bucket</code> (or <code>remote:</code> for the <code>lsf</code> command.) You may put subdirectories in too, e.g. <code>remote:bucket/path/to/dir</code>.</p>
<p><code>rclone about</code> is not supported by the rclone Tardigrade backend. Backends without this capability cannot determine free space for an rclone mount or use policy <code>mfs</code> (most free space) as a member of an rclone union remote.</p>
<p>See <ahref="https://rclone.org/overview/#optional-features">List of backends that do not support rclone about</a> See <ahref="https://rclone.org/commands/rclone_about/">rclone about</a></p>
<p>During the initial setup with <code>rclone config</code> you will specify the upstream remotes as a space separated list. The upstream remotes can either be a local paths or other remotes.</p>
<p>Attribute <code>:ro</code> and <code>:nc</code> can be attach to the end of path to tag the remote as <strong>read only</strong> or <strong>no create</strong>, e.g. <code>remote:directory/subdirectory:ro</code> or <code>remote:directory/subdirectory:nc</code>.</p>
<p>Subfolders can be used in upstream remotes. Assume a union remote named <code>backup</code> with the remotes <code>mydrive:private/backup</code>. Invoking <code>rclone mkdir backup:desktop</code> is exactly the same as invoking <code>rclone mkdir mydrive:private/backup/desktop</code>.</p>
<p>There will be no special handling of paths containing <code>..</code> segments. Invoking <code>rclone mkdir backup:../desktop</code> is exactly the same as invoking <code>rclone mkdir mydrive:private/backup/../desktop</code>.</p>
<p>The behavior of union backend is inspired by <ahref="https://github.com/trapexit/mergerfs">trapexit/mergerfs</a>. All functions are grouped into 3 categories: <strong>action</strong>, <strong>create</strong> and <strong>search</strong>. These functions and categories can be assigned a policy which dictates what file or directory is chosen when performing that behavior. Any policy can be assigned to a function or category though some may not be very useful in practice. For instance: <strong>rand</strong> (random) may be useful for file creation (create) but could lead to very odd behavior if used for <code>delete</code> if there were more than one copy of the file.</p>
<p>Policies, as described below, are of two basic types. <code>path preserving</code> and <code>non-path preserving</code>.</p>
<p>All policies which start with <code>ep</code> (<strong>epff</strong>, <strong>eplfs</strong>, <strong>eplus</strong>, <strong>epmfs</strong>, <strong>eprand</strong>) are <code>path preserving</code>. <code>ep</code> stands for <code>existing path</code>.</p>
<p>A path preserving policy will only consider upstreams where the relative path being accessed already exists.</p>
<p>When using non-path preserving policies paths will be created in target upstreams as necessary.</p>
<p>Some policies rely on quota information. These policies should be used only if your upstreams support the respective quota fields.</p>
<table>
<thead>
<trclass="header">
<th>Policy</th>
<th>Required Field</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>lfs, eplfs</td>
<td>Free</td>
</tr>
<trclass="even">
<td>mfs, epmfs</td>
<td>Free</td>
</tr>
<trclass="odd">
<td>lus, eplus</td>
<td>Used</td>
</tr>
<trclass="even">
<td>lno, eplno</td>
<td>Objects</td>
</tr>
</tbody>
</table>
<p>To check if your upstream supports the field, run <code>rclone about remote: [flags]</code> and see if the required field exists.</p>
<h4id="filters-1">Filters</h4>
<p>Policies basically search upstream remotes and create a list of files / paths for functions to work on. The policy is responsible for filtering and sorting. The policy type defines the sorting but filtering is mostly uniform as described below.</p>
<p>The policies definition are inspired by <ahref="https://github.com/trapexit/mergerfs">trapexit/mergerfs</a> but not exactly the same. Some policy definition could be different due to the much larger latency of remote file systems.</p>
<table>
<colgroup>
<colstyle="width: 23%"/>
<colstyle="width: 76%"/>
</colgroup>
<thead>
<trclass="header">
<th>Policy</th>
<th>Description</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>all</td>
<td>Search category: same as <strong>epall</strong>. Action category: same as <strong>epall</strong>. Create category: act on all upstreams.</td>
</tr>
<trclass="even">
<td>epall (existing path, all)</td>
<td>Search category: Given this order configured, act on the first one found where the relative path exists. Action category: apply to all found. Create category: act on all upstreams where the relative path exists.</td>
</tr>
<trclass="odd">
<td>epff (existing path, first found)</td>
<td>Act on the first one found, by the time upstreams reply, where the relative path exists.</td>
</tr>
<trclass="even">
<td>eplfs (existing path, least free space)</td>
<td>Of all the upstreams on which the relative path exists choose the one with the least free space.</td>
</tr>
<trclass="odd">
<td>eplus (existing path, least used space)</td>
<td>Of all the upstreams on which the relative path exists choose the one with the least used space.</td>
</tr>
<trclass="even">
<td>eplno (existing path, least number of objects)</td>
<td>Of all the upstreams on which the relative path exists choose the one with the least number of objects.</td>
</tr>
<trclass="odd">
<td>epmfs (existing path, most free space)</td>
<td>Of all the upstreams on which the relative path exists choose the one with the most free space.</td>
</tr>
<trclass="even">
<td>eprand (existing path, random)</td>
<td>Calls <strong>epall</strong> and then randomizes. Returns only one upstream.</td>
</tr>
<trclass="odd">
<td>ff (first found)</td>
<td>Search category: same as <strong>epff</strong>. Action category: same as <strong>epff</strong>. Create category: Act on the first one found by the time upstreams reply.</td>
</tr>
<trclass="even">
<td>lfs (least free space)</td>
<td>Search category: same as <strong>eplfs</strong>. Action category: same as <strong>eplfs</strong>. Create category: Pick the upstream with the least available free space.</td>
</tr>
<trclass="odd">
<td>lus (least used space)</td>
<td>Search category: same as <strong>eplus</strong>. Action category: same as <strong>eplus</strong>. Create category: Pick the upstream with the least used space.</td>
</tr>
<trclass="even">
<td>lno (least number of objects)</td>
<td>Search category: same as <strong>eplno</strong>. Action category: same as <strong>eplno</strong>. Create category: Pick the upstream with the least number of objects.</td>
</tr>
<trclass="odd">
<td>mfs (most free space)</td>
<td>Search category: same as <strong>epmfs</strong>. Action category: same as <strong>epmfs</strong>. Create category: Pick the upstream with the most available free space.</td>
</tr>
<trclass="even">
<td>newest</td>
<td>Pick the file / directory with the largest mtime.</td>
</tr>
<trclass="odd">
<td>rand (random)</td>
<td>Calls <strong>all</strong> and then randomizes. Returns only one upstream.</td>
<p>To configure the WebDAV remote you will need to have a URL for it, and a username and password. If you know what kind of system you are connecting to then rclone can enable extra features.</p>
<p>Here is an example of how to make a remote called <code>remote</code>. First run:</p>
<pre><code> rclone config</code></pre>
<p>This will guide you through an interactive setup process:</p>
<pre><code>No remotes found - make a new one
n) New remote
s) Set configuration password
q) Quit config
n/s/q> n
name> remote
Type of storage to configure.
Choose a number from below, or type in your own value
<p>Likewise plain WebDAV does not support hashes, however when used with Owncloud or Nextcloud rclone will support SHA1 and MD5 hashes. Depending on the exact version of Owncloud or Nextcloud hashes may appear on all objects, or only on objects which had a hash uploaded with them.</p>
<p>See: the <ahref="https://rclone.org/overview/#encoding">encoding section in the overview</a> for more info.</p>
<p>Default encoding is Slash,LtGt,DoubleQuote,Colon,Question,Asterisk,Pipe,Hash,Percent,BackSlash,Del,Ctl,LeftSpace,LeftTilde,RightSpace,RightPeriod,InvalidUtf8 for sharepoint-ntlm or identity otherwise.</p>
<p>Click on the settings cog in the bottom right of the page and this will show the WebDAV URL that rclone needs in the config step. It will look something like <code>https://example.com/remote.php/webdav/</code>.</p>
<p>Owncloud supports modified times using the <code>X-OC-Mtime</code> header.</p>
<p>This is configured in an identical way to Owncloud. Note that Nextcloud initially did not support streaming of files (<code>rcat</code>) whereas Owncloud did, but <ahref="https://github.com/nextcloud/nextcloud-snap/issues/365">this</a> seems to be fixed as of 2020-11-27 (tested with rclone v1.53.1 and Nextcloud Server v19).</p>
<p>Rclone can be used with Sharepoint provided by OneDrive for Business or Office365 Education Accounts. This feature is only needed for a few of these Accounts, mostly Office365 Education ones. These accounts are sometimes not verified by the domain owner <ahref="https://github.com/rclone/rclone/issues/1975">github#1975</a></p>
<p>This means that these accounts can't be added using the official API (other Accounts should work with the "onedrive" option). However, it is possible to access them using webdav.</p>
<p>To use a sharepoint remote with rclone, add it like this: First, you need to get your remote's URL:</p>
<li>Go <ahref="https://onedrive.live.com/about/en-us/signin/">here</a> to open your OneDrive or to sign in</li>
<li>Now take a look at your address bar, the URL should look like this: <code>https://[YOUR-DOMAIN]-my.sharepoint.com/personal/[YOUR-EMAIL]/_layouts/15/onedrive.aspx</code></li>
<p>You'll only need this URL up to the email address. After that, you'll most likely want to add "/Documents". That subdirectory contains the actual data stored on your OneDrive.</p>
<p>Add the remote to rclone like this: Configure the <code>url</code> as <code>https://[YOUR-DOMAIN]-my.sharepoint.com/personal/[YOUR-EMAIL]/Documents</code> and use your normal account email and password for <code>user</code> and <code>pass</code>. If you have 2FA enabled, you have to generate an app password. Set the <code>vendor</code> to <code>sharepoint</code>.</p>
<h3id="sharepoint-with-ntlm-authentication">Sharepoint with NTLM Authentication</h3>
<p>Use this option in case your (hosted) Sharepoint is not tied to OneDrive accounts and uses NTLM authentication.</p>
<p>To get the <code>url</code> configuration, similarly to the above, first navigate to the desired directory in your browser to get the URL, then strip everything after the name of the opened directory.</p>
<p>Example: If the URL is: https://example.sharepoint.com/sites/12345/Documents/Forms/AllItems.aspx</p>
<p>The configuration to use would be: https://example.sharepoint.com/sites/12345/Documents</p>
<p>Set the <code>vendor</code> to <code>sharepoint-ntlm</code>.</p>
<p>NTLM uses domain and user name combination for authentication, set <code>user</code> to <code>DOMAIN\username</code>.</p>
<p>As SharePoint does some special things with uploaded documents, you won't be able to use the documents size or the documents hash to compare if a file has been changed since the upload / which file is newer.</p>
<p>For Rclone calls copying files (especially Office files such as .docx, .xlsx, etc.) from/to SharePoint (like copy, sync, etc.), you should append these flags to ensure Rclone uses the "Last Modified" datetime property to compare your documents:</p>
<p>dCache is a storage system that supports many protocols and authentication/authorisation schemes. For WebDAV clients, it allows users to authenticate with username and password (BASIC), X.509, Kerberos, and various bearer tokens, including <ahref="https://www.dcache.org/manuals/workshop-2017-05-29-Umea/000-Final/anupam_macaroons_v02.pdf">Macaroons</a> and <ahref="https://en.wikipedia.org/wiki/OpenID_Connect">OpenID-Connect</a> access tokens.</p>
<p>Configure as normal using the <code>other</code> type. Don't enter a username or password, instead enter your Macaroon as the <code>bearer_token</code>.</p>
<p>There is a <ahref="https://github.com/sara-nl/GridScripts/blob/master/get-macaroon">script</a> that obtains a Macaroon from a dCache WebDAV endpoint, and creates an rclone config file.</p>
<p>Macaroons may also be obtained from the dCacheView web-browser/JavaScript client that comes with dCache.</p>
<h3id="openid-connect">OpenID-Connect</h3>
<p>dCache also supports authenticating with OpenID-Connect access tokens. OpenID-Connect is a protocol (based on OAuth 2.0) that allows services to identify users who have authenticated with some central service.</p>
<p>Support for OpenID-Connect in rclone is currently achieved using another software package called <ahref="https://github.com/indigo-dc/oidc-agent">oidc-agent</a>. This is a command-line tool that facilitates obtaining an access token. Once installed and configured, an access token is obtained by running the <code>oidc-token</code> command. The following example shows a (shortened) access token obtained from the <em>XDC</em> OIDC Provider.</p>
<pre><code>paul@celebrimbor:~$ oidc-token XDC
eyJraWQ[...]QFXDt0
paul@celebrimbor:~$</code></pre>
<p><strong>Note</strong> Before the <code>oidc-token</code> command will work, the refresh token must be loaded into the oidc agent. This is done with the <code>oidc-add</code> command (e.g., <code>oidc-add XDC</code>). This is typically done once per login session. Full details on this and how to register oidc-agent with your OIDC Provider are provided in the <ahref="https://indigo-dc.gitbooks.io/oidc-agent/">oidc-agent documentation</a>.</p>
<p>The rclone <code>bearer_token_command</code> configuration option is used to fetch the access token from oidc-agent.</p>
<p>Configure as a normal WebDAV endpoint, using the 'other' vendor, leaving the username and password empty. When prompted, choose to edit the advanced config and enter the command to get a bearer token (e.g., <code>oidc-agent XDC</code>).</p>
<p>See the <ahref="https://rclone.org/remote_setup/">remote setup docs</a> for how to set it up on a machine with no Internet browser available.</p>
<p>Note that rclone runs a webserver on your local machine to collect the token as returned from Yandex Disk. This only runs from the moment it opens your browser to the moment you get back the verification code. This is on <code>http://127.0.0.1:53682/</code> and this it may require you to unblock it temporarily if you are running a host firewall.</p>
<p>Once configured you can then use <code>rclone</code> like this,</p>
<p>Modified times are supported and are stored accurate to 1 ns in custom metadata called <code>rclone_modified</code> in RFC3339 with nanoseconds format.</p>
<h3id="md5-checksums">MD5 checksums</h3>
<p>MD5 checksums are natively supported by Yandex Disk.</p>
<p>If you wish to empty your trash you can use the <code>rclone cleanup remote:</code> command which will permanently delete all your trashed files. This command does not take any path arguments.</p>
<p>To view your current quota you can use the <code>rclone about remote:</code> command which will display your usage limit (quota) and the current usage.</p>
<p>When uploading very large files (bigger than about 5GB) you will need to increase the <code>--timeout</code> parameter. This is because Yandex pauses (perhaps to calculate the MD5SUM for the entire file) before returning confirmation that the file has been uploaded. The default handling of timeouts in rclone is to assume a 5 minute pause is an error and close the connection - you'll see <code>net/http: timeout awaiting response headers</code> errors in the logs if this is happening. Setting the timeout to twice the max size of file in GB should be enough, so if you want to upload a 30GB file set a timeout of <code>2 * 30 = 60m</code>, that is <code>--timeout 60m</code>.</p>
<p>Rclone runs a webserver on your local computer to collect the authorization token from Zoho Workdrive. This is only from the moment your browser is opened until the token is returned. The webserver runs on <code>http://127.0.0.1:53682/</code>. If local port <code>53682</code> is protected by a firewall you may need to temporarily unblock the firewall to complete authorization.</p>
<p>Only control characters and invalid UTF-8 are replaced. In addition most Unicode full-width characters are not supported at all and will be removed from filenames during upload.</p>
<h3id="standard-options-37">Standard Options</h3>
<p>Here are the standard options specific to zoho (Zoho).</p>
<p>Will sync <code>/home/source</code> to <code>/tmp/destination</code>.</p>
<p>For consistencies sake one can also configure a remote of type <code>local</code> in the config file, and access the local filesystem using rclone remote paths, e.g. <code>remote:path/to/wherever</code>, but it is probably easier not to.</p>
<p>Rclone reads and writes the modified time using an accuracy determined by the OS. Typically this is 1ns on Linux, 10 ns on Windows and 1 Second on OS X.</p>
<p>There is a bit more uncertainty in the Linux world, but new distributions will have UTF-8 encoded files names. If you are using an old Linux filesystem with non UTF-8 file names (e.g. latin1) then you can use the <code>convmv</code> tool to convert the filesystem to UTF-8. This tool is available in most distributions' package managers.</p>
<p>If an invalid (non-UTF8) filename is read, the invalid characters will be replaced with a quoted representation of the invalid bytes. The name <code>gro\xdf</code> will be transferred as <code>gro‛DF</code>. <code>rclone</code> will emit a debug message in this case (use <code>-v</code> to see), e.g.</p>
<p>On non Windows platforms the following characters are replaced when handling file names.</p>
<table>
<thead>
<trclass="header">
<th>Character</th>
<thstyle="text-align: center;">Value</th>
<thstyle="text-align: center;">Replacement</th>
</tr>
</thead>
<tbody>
<trclass="odd">
<td>NUL</td>
<tdstyle="text-align: center;">0x00</td>
<tdstyle="text-align: center;">␀</td>
</tr>
<trclass="even">
<td>/</td>
<tdstyle="text-align: center;">0x2F</td>
<tdstyle="text-align: center;">／</td>
</tr>
</tbody>
</table>
<p>When running on Windows the following characters are replaced. This list is based on the <ahref="https://docs.microsoft.com/de-de/windows/desktop/FileIO/naming-a-file#naming-conventions">Windows file naming conventions</a>.</p>
<p>On Windows there are many ways of specifying a path to a file system resource. Both absolute paths like <code>C:\path\to\wherever</code>, and relative paths like <code>..\wherever</code> can be used, and path separator can be either <code>\</code> (as in <code>C:\path\to\wherever</code>) or <code>/</code> (as in <code>C:/path/to/wherever</code>). Length of these paths are limited to 259 characters for files and 247 characters for directories, but there is an alternative extended-length path format increasing the limit to (approximately) 32,767 characters. This format requires absolute paths and the use of prefix <code>\\?\</code>, e.g. <code>\\?\D:\some\very\long\path</code>. For convenience rclone will automatically convert regular paths into the corresponding extended-length paths, so in most cases you do not have to worry about this (read more <ahref="#long-paths">below</a>).</p>
<p>Note that Windows supports using the same prefix <code>\\?\</code> to specify path to volumes identified by their GUID, e.g. <code>\\?\Volume{b75e2c83-0000-0000-0000-602f00000000}\some\path</code>. This is <em>not</em> supported in rclone, due to an <ahref="https://github.com/golang/go/issues/39785">issue</a> in go.</p>
<h4id="long-paths">Long paths</h4>
<p>Rclone handles long paths automatically, by converting all paths to <ahref="https://docs.microsoft.com/en-us/windows/win32/fileio/maximum-file-path-limitation">extended-length path format</a>, which allows paths up to 32,767 characters.</p>
<p>This conversion will ensure paths are absolute and prefix them with the <code>\\?\</code>. This is why you will see that your paths, for instance <code>.\files</code> is shown as path <code>\\?\C:\files</code> in the output, and <code>\\server\share</code> as <code>\\?\UNC\server\share</code>.</p>
<p>However, in rare cases this may cause problems with buggy file system drivers like <ahref="https://github.com/rclone/rclone/issues/261">EncFS</a>. To disable UNC conversion globally, add this to your <code>.rclone.conf</code> file:</p>
<p>This will use UNC paths on <code>c:\src</code> but not on <code>z:\dst</code>. Of course this will cause problems if the absolute path length of a file exceeds 259 characters on z, so only use this option if you have to.</p>
<p>If you supply <code>--copy-links</code> or <code>-L</code> then rclone will follow the symlink and copy the pointed to file or directory. Note that this flag is incompatible with <code>-links</code> / <code>-l</code>.</p>
<p>If you supply this flag then rclone will copy symbolic links from the local storage, and store them as text files, with a '.rclonelink' suffix in the remote storage.</p>
<p>Normally rclone will recurse through filesystems as mounted.</p>
<p>However if you set <code>--one-file-system</code> or <code>-x</code> this tells rclone to stay in the filesystem specified by the root and not to recurse into different file systems.</p>
<p>Using <code>rclone --one-file-system copy root remote:</code> will only copy <code>file1</code> and <code>file2</code>. Eg</p>
<pre><code>$ rclone -q --one-file-system ls root
0 file1
0 file2</code></pre>
<pre><code>$ rclone -q ls root
0 disk1/file3
0 disk2/file4
0 file1
0 file2</code></pre>
<p><strong>NB</strong> Rclone (like most unix tools such as <code>du</code>, <code>rsync</code> and <code>tar</code>) treats a bind mount to the same device as being on the same filesystem.</p>
<p>Follow symlinks and copy the pointed to item.</p>
<ul>
<li>Config: copy_links</li>
<li>Env Var: RCLONE_LOCAL_COPY_LINKS</li>
<li>Type: bool</li>
<li>Default: false</li>
</ul>
<h4id="links--l-1">--links / -l</h4>
<p>Translate symlinks to/from regular files with a '.rclonelink' extension</p>
<ul>
<li>Config: links</li>
<li>Env Var: RCLONE_LOCAL_LINKS</li>
<li>Type: bool</li>
<li>Default: false</li>
</ul>
<h4id="skip-links">--skip-links</h4>
<p>Don't warn about skipped symlinks. This flag disables warning messages on skipped symlinks or junction points, as you explicitly acknowledge that they should be skipped.</p>
<p>Assume the Stat size of links is zero (and read them instead)</p>
<p>On some virtual filesystems (such ash LucidLink), reading a link size via a Stat call always returns 0. However, on unix it reads as the length of the text in the link. This may cause errors like this when syncing:</p>
<pre><code>Failed to copy: corrupted on transfer: sizes differ 0 vs 13</code></pre>
<p>Setting this flag causes rclone to read the link and use that as the size of the link instead of 0 which in most cases fixes the problem.</p>
<p>Don't apply unicode normalization to paths and filenames (Deprecated)</p>
<p>This flag is deprecated now. Rclone no longer normalizes unicode file names, but it compares them with unicode normalization in the sync routine instead.</p>
<p>Don't check to see if the files change during upload</p>
<p>Normally rclone checks the size and modification time of files as they are being uploaded and aborts with a message which starts "can't copy - source file is being updated" if the file changes during upload.</p>
<p>However on some file systems this modification time check may fail (e.g. <ahref="https://github.com/rclone/rclone/issues/2206">Glusterfs #2206</a>) so this check can be disabled with this flag.</p>
<p>If this flag is set, rclone will use its best efforts to transfer a file which is being updated. If the file is only having things appended to it (e.g. a log) then rclone will transfer the log file with the size it had the first time rclone saw it.</p>
<p>If the file is being modified throughout (not just appended to) then the transfer may fail with a hash check failure.</p>
<p>In detail, once the file has had stat() called on it for the first time we:</p>
<ul>
<li>Only transfer the size that stat gave</li>
<li>Only checksum the size that stat gave</li>
<li><p>Don't update the stat info for the file</p></li>
<p>Force the filesystem to report itself as case sensitive.</p>
<p>Normally the local backend declares itself as case insensitive on Windows/macOS and case sensitive for everything else. Use this flag to override the default choice.</p>
<p>Force the filesystem to report itself as case insensitive</p>
<p>Normally the local backend declares itself as case insensitive on Windows/macOS and case sensitive for everything else. Use this flag to override the default choice.</p>
<p>Disable preallocation of disk space for transferred files</p>
<p>Preallocation of disk space helps prevent filesystem fragmentation. However, some virtual filesystem layers (such as Google Drive File Stream) may incorrectly set the actual file size equal to the preallocated space, causing checksum and file size checks to fail. Use this flag to disable preallocation.</p>
<p>Disable sparse files for multi-thread downloads</p>
<p>On Windows platforms rclone will make sparse files when doing multi-thread downloads. This avoids long pauses on large files where the OS zeros the file. However sparse files may be undesirable as they cause disk fragmentation and can be slow to work with.</p>
<p>Normally rclone updates modification time of files after they are done uploading. This can cause permissions issues on Linux platforms when the user rclone is running as does not own the file uploaded, such as when copying to a CIFS mount owned by another user. If this option is enabled, rclone will no longer update the modtime after copying a file.</p>
<ul>
<li>Config: no_set_modtime</li>
<li>Env Var: RCLONE_LOCAL_NO_SET_MODTIME</li>
<li>Type: bool</li>
<li>Default: false</li>
</ul>
<h4id="local-encoding">--local-encoding</h4>
<p>This sets the encoding for the backend.</p>
<p>See: the <ahref="https://rclone.org/overview/#encoding">encoding section in the overview</a> for more info.</p>
<ul>
<li>Config: encoding</li>
<li>Env Var: RCLONE_LOCAL_ENCODING</li>
<li>Type: MultiEncoder</li>
<li>Default: Slash,Dot</li>
</ul>
<h3id="backend-commands-4">Backend commands</h3>
<p>Here are the commands specific to the local backend.</p>
<li>The <ahref="https://rclone.org/commands/rclone_mount/#vfs-virtual-file-system">VFS layer</a> was heavily reworked for this release - see below for more details</li>
<li>Interactive mode <ahref="https://rclone.org/docs/#interactive">-i/--interactive</a> for destructive operations (fishbullet)</li>
<li>Add <ahref="https://rclone.org/docs/#bwlimit-file-bandwidth-spec">--bwlimit-file</a> flag to limit speeds of individual file transfers (Nick Craig-Wood)</li>
<li>Transfers are sorted by start time in the stats and progress output (Max Sum)</li>
<li>Make sure backends expand <code>~</code> and environment vars in file names they use (Nick Craig-Wood)</li>
<li>Add <ahref="https://rclone.org/docs/#refresh-times">--refresh-times</a> flag to set modtimes on hashless backends (Nick Craig-Wood)</li>
<li>build
<ul>
<li>Remove vendor directory in favour of Go modules (Nick Craig-Wood)</li>
<li>Build with go1.15.x by default (Nick Craig-Wood)</li>
<li>Drop macOS 386 build as it is no longer supported by go1.15 (Nick Craig-Wood)</li>
<li>Add ARMv7 to the supported builds (Nick Craig-Wood)</li>
<li>Enable <code>rclone cmount</code> on macOS (Nick Craig-Wood)</li>
<li>Make rclone build with gccgo (Nick Craig-Wood)</li>
<li>Make rclone build with wasm (Nick Craig-Wood)</li>
<li>Change beta numbering to be semver compatible (Nick Craig-Wood)</li>
<li>Add file properties and icon to Windows executable (albertony)</li>
<li>Add experimental interface for integrating rclone into browsers (Nick Craig-Wood)</li>
</ul></li>
<li>lib: Add file name compression (Klaus Post)</li>
<li>rc
<ul>
<li>Allow installation and use of plugins and test plugins with rclone-webui (Chaitanya Bankanhal)</li>
<li>Add reverse proxy pluginsHandler for serving plugins (Chaitanya Bankanhal)</li>
<li>Add <code>mount/listmounts</code> option for listing current mounts (Chaitanya Bankanhal)</li>
<li>Add <code>operations/uploadfile</code> to upload a file through rc using encoding multipart/form-data (Chaitanya Bankanhal)</li>
<li><code>rclone touch</code>: Add ability to set nanosecond resolution times (Nick Craig-Wood)</li>
<li><code>rclone tree</code>: Remove <code>-i</code> shorthand for <code>--noindent</code> as it conflicts with <code>-i</code>/<code>--interactive</code> (Nick Craig-Wood)</li>
</ul></li>
<li>Bug Fixes
<ul>
<li>accounting
<ul>
<li>Fix documentation for <code>speed</code>/<code>speedAvg</code> (Nick Craig-Wood)</li>
<li>Fix elapsed time not show actual time since beginning (Chaitanya Bankanhal)</li>
<li>Fix deadlock in stats printing (Nick Craig-Wood)</li>
</ul></li>
<li>build
<ul>
<li>Fix file handle leak in GitHub release tool (Garrett Squire)</li>
<li>Change maximum leaf name length to 1024 bytes (Nick Craig-Wood)</li>
</ul></li>
<li>VFS
<ul>
<li>Add <code>--vfs-read-wait</code> and <code>--vfs-write-wait</code> flags to control time waiting for a sequential read/write (Nick Craig-Wood)</li>
<li>Change default <code>--vfs-read-wait</code> to 20ms (it was 5ms and not configurable) (Nick Craig-Wood)</li>
<li>Make <code>df</code> output more consistent on a rclone mount. (Yves G)</li>
<li>Report 1PB free for unknown disk sizes (Nick Craig-Wood)</li>
<li>Fix race condition caused by unlocked reading of Dir.path (Nick Craig-Wood)</li>
<li>Make File lock and Dir lock not overlap to avoid deadlock (Nick Craig-Wood)</li>
<li>Implement lock ordering between File and Dir to eliminate deadlocks (Nick Craig-Wood)</li>
<li>Factor the vfs cache into its own package (Nick Craig-Wood)</li>
<li>Pin the Fs in use in the Fs cache (Nick Craig-Wood)</li>
<li>Add SetSys() methods to Node to allow caching stuff on a node (Nick Craig-Wood)</li>
<li>Ignore file not found errors from Hash in Read.Release (Nick Craig-Wood)</li>
<li>Fix hang in read wait code (Nick Craig-Wood)</li>
</ul></li>
<li>Local
<ul>
<li>Speed up multi thread downloads by using sparse files on Windows (Nick Craig-Wood)</li>
<li>Implement <code>--local-no-sparse</code> flag for disabling sparse files (Nick Craig-Wood)</li>
<li>Implement <code>rclone backend noop</code> for testing purposes (Nick Craig-Wood)</li>
<li>See the <ahref="https://rclone.org/overview/#restricted-filenames">restricted file name docs</a> for more info and the <ahref="/local/#filenames">local backend docs</a>.</li>
<li>Some file names may look different in rclone if you are using any control characters in names or <ahref="https://en.wikipedia.org/wiki/Halfwidth_and_Fullwidth_Forms_(Unicode_block)">unicode FULLWIDTH symbols</a>.</li>
</ul></li>
<li>build
<ul>
<li>Update to use go1.13 for the build (Nick Craig-Wood)</li>
<li>Drop support for go1.9 (Nick Craig-Wood)</li>
<li>Build rclone with GitHub actions (Nick Craig-Wood)</li>
<li>Convert python scripts to python3 (Nick Craig-Wood)</li>
<li>Swap Azure/go-ansiterm for mattn/go-colorable (Nick Craig-Wood)</li>
<li>Dockerfile fixes (Matei David)</li>
<li>Add <ahref="https://github.com/rclone/rclone/blob/master/CONTRIBUTING.md#writing-a-plugin">plugin support</a> for backends and commands (Richard Patel)</li>
</ul></li>
<li>config
<ul>
<li>Use alternating Red/Green in config to make more obvious (Nick Craig-Wood)</li>
</ul></li>
<li>contrib
<ul>
<li>Add sample DLNA server Docker Compose manifest. (pataquets)</li>
<li>Add sample WebDAV server Docker Compose manifest. (pataquets)</li>
</ul></li>
<li>copyurl
<ul>
<li>Add <code>--auto-filename</code> flag for using file name from URL in destination path (Denis)</li>
<li>Logging revamped to be more inline with rsync - now much quieter * -v only shows transfers * -vv is for full debug * --syslog to log to syslog on capable platforms</li>
<li>Implement --backup-dir and --suffix</li>
<li>Implement --track-renames (initial implementation by Bjørn Erik Pedersen)</li>
<li>use official AWS SDK from github.com/aws/aws-sdk-go</li>
<li><strong>NB</strong> will most likely require you to delete and recreate remote</li>
<li>enable multipart upload which enables files > 5GB</li>
<li>tested with Ceph / RadosGW / S3 emulation</li>
<li>many thanks to Sam Liston and Brian Haymore at the <ahref="https://www.chpc.utah.edu/">Utah Center for High Performance Computing</a> for a Ceph test account</li>
<p>Rclone doesn't currently preserve the timestamps of directories. This is because rclone only really considers objects when syncing.</p>
<h3id="rclone-struggles-with-millions-of-files-in-a-directorybucket">Rclone struggles with millions of files in a directory/bucket</h3>
<p>Currently rclone loads each directory/bucket entirely into memory before using it. Since each rclone object takes 0.5k-1k of memory this can take a very long time and use a large amount of memory.</p>
<p>Millions of files in a directory tends to occur on bucket-based remotes (e.g. S3 buckets) since those remotes do not segregate subdirectories within the bucket.</p>
<p>Bucket based remotes (e.g. S3/GCS/Swift/B2) do not have a concept of directories. Rclone therefore cannot create directories in them which means that empty directories on a bucket based remote will tend to disappear.</p>
<p>Some software creates empty keys ending in <code>/</code> as directory markers. Rclone doesn't do this as it potentially creates more objects and costs more. This ability may be added in the future (probably via a flag/option).</p>
<p>Sure! Rclone stores all of its config in a single file. If you want to find this file, run <code>rclone config file</code> which will tell you where it is.</p>
<p>The file names you upload from Server A and Server B should be different in this case, otherwise some file systems (e.g. Drive) may make duplicates.</p>
<p>Rclone stores each file you transfer as a native object on the remote cloud storage system. This means that you can see the files you upload as expected using alternative access methods (e.g. using the Google Drive web interface). There is a 1:1 mapping between files on your hard disk and objects created in the cloud storage system.</p>
<p>Cloud storage systems (at least none I've come across yet) don't support partially uploading an object. You can't take an existing object, and change some bytes in the middle of it.</p>
<p>It would be possible to make a sync system which stored binary diffs instead of whole objects like rclone does, but that would break the 1:1 mapping of files on your hard disk to objects in the remote cloud storage system.</p>
<p>All the cloud storage systems support partial downloads of content, so it would be possible to make partial downloads work. However to make this work efficiently this would require storing a significant amount of metadata, which breaks the desired 1:1 mapping of files to objects.</p>
<h3id="can-rclone-do-bi-directional-sync">Can rclone do bi-directional sync?</h3>
<p>No, not at present. rclone only does uni-directional sync from A -> B. It may do in the future though since it has all the primitives - it just requires writing the algorithm to do it.</p>
<h3id="can-i-use-rclone-with-an-http-proxy">Can I use rclone with an HTTP proxy?</h3>
<p>Yes. rclone will follow the standard environment variables for proxies, similar to cURL and other programs.</p>
<p>In general the variables are called <code>http_proxy</code> (for services reached over <code>http</code>) and <code>https_proxy</code> (for services reached over <code>https</code>). Most public services will be using <code>https</code>, but you may wish to set both.</p>
<p>The content of the variable is <code>protocol://server:port</code>. The protocol value is the one used to talk to the proxy server, itself, and is commonly either <code>http</code> or <code>socks5</code>.</p>
<p>Slightly annoyingly, there is no <em>standard</em> for the name; some applications may use <code>http_proxy</code> but another one <code>HTTP_PROXY</code>. The <code>Go</code> libraries used by <code>rclone</code> will try both variations, but you may wish to set all possibilities. So, on Linux, you may end up with code similar to</p>
<p>The <code>NO_PROXY</code> allows you to disable the proxy for specific hosts. Hosts must be comma separated, and can contain domains or parts. For instance "foo.com" also matches "bar.foo.com".</p>
<h3id="rclone-gives-x509-failed-to-load-system-roots-and-no-roots-provided-error">Rclone gives x509: failed to load system roots and no roots provided error</h3>
<p>This means that <code>rclone</code> can't find the SSL root certificates. Likely you are running <code>rclone</code> on a NAS with a cut-down Linux OS, or possibly on Solaris.</p>
<p>The two environment variables <code>SSL_CERT_FILE</code> and <code>SSL_CERT_DIR</code>, mentioned in the <ahref="https://godoc.org/crypto/x509">x509 package</a>, provide an additional way to provide the SSL root certificates.</p>
<h3id="rclone-gives-failed-to-load-config-file-function-not-implemented-error">Rclone gives Failed to load config file: function not implemented error</h3>
<p>Likely this means that you are running rclone on Linux version not supported by the go runtime, ie earlier than version 2.6.23.</p>
<p>See the <ahref="https://golang.org/doc/install">system requirements section in the go install docs</a> for full details.</p>
<p>This is caused by uploading these files from a Windows computer which hasn't got the Microsoft Office suite installed. The easiest way to fix is to install the Word viewer and the Microsoft Office Compatibility Pack for Word, Excel, and PowerPoint 2007 and later versions' file formats</p>
<h3id="tcp-lookup-some.domain.com-no-such-host">tcp lookup some.domain.com no such host</h3>
<p>This happens when rclone cannot resolve a domain. Please check that your DNS setup is generally working, e.g.</p>
<pre><code># both should print a long list of possible IP addresses
dig www.googleapis.com # resolve using your default DNS
dig www.googleapis.com @8.8.8.8 # resolve with Google's DNS server</code></pre>
<p>If you are using <code>systemd-resolved</code> (default on Arch Linux), ensure it is at version 233 or higher. Previous releases contain a bug which causes not all domains to be resolved properly.</p>
<p>Additionally with the <code>GODEBUG=netdns=</code> environment variable the Go resolver decision can be influenced. This also allows to resolve certain issues with DNS resolution. See the <ahref="https://golang.org/pkg/net/#hdr-Name_Resolution">name resolution section in the go docs</a>.</p>
<h3id="the-total-size-reported-in-the-stats-for-a-sync-is-wrong-and-keeps-changing">The total size reported in the stats for a sync is wrong and keeps changing</h3>
<p>It is likely you have more than 10,000 files that need to be synced. By default rclone only gets 10,000 files ahead in a sync so as not to use up too much memory. You can change this default with the <ahref="https://rclone.org/docs/#max-backlog-n">--max-backlog</a> flag.</p>
<h3id="rclone-is-using-too-much-memory-or-appears-to-have-a-memory-leak">Rclone is using too much memory or appears to have a memory leak</h3>
<p>Rclone is written in Go which uses a garbage collector. The default settings for the garbage collector mean that it runs when the heap size has doubled.</p>
<p>However it is possible to tune the garbage collector to use less memory by <ahref="https://dave.cheney.net/tag/gogc">setting GOGC</a> to a lower value, say <code>export GOGC=20</code>. This will make the garbage collector work harder, reducing memory size at the expense of CPU usage.</p>
<p>The most common cause of rclone using lots of memory is a single directory with thousands or millions of files in. Rclone has to load this entirely into memory as rclone objects. Each rclone object takes 0.5k-1k of memory.</p>
<p>{{< rem <code>email addresses removed from here need to be addeed to bin/.ignore-emails to make sure update-authors.py doesn't immediately put them back in again.</code>>}}</p>
<p>Or if all else fails or you want to ask something private or confidential email <ahref="mailto:nick@craig-wood.com">Nick Craig-Wood</a>. Please don't email me requests for help - those are better directed to the forum. Thanks!</p>
