Rclone systemd user service

Rclone systemd service.md

rclone systemd service

Preparation

This service will use the same remote name you specified when using rclone config create. If you haven't done that yet, do so now.

Next, create the mountpoint for your remote. The service uses the location ~/mnt/<remote> by default.

mkdir ~/mnt/dropbox

The --allow-other option is required in order to work in many desktop environments. This flag must be enabled by adding user_allow_other to /etc/fuse.conf. If you aren't using a desktop environment, such as on a server, this option can be omitted.

Adding the service

Save the [email protected] file in ~/.config/systemd/user/ Make sure you include the @. This is required to work.

As your normal user, run:

systemctl --user daemon-reload

Using the service

You can now start/enable each remote by using rclone@<remote>

systemctl --user enable --now rclone@dropbox

[email protected]

# User service for Rclone mounting
#
# Place in ~/.config/systemd/user/
# File must include the '@' (ex [email protected])
# As your normal user, run 
#   systemctl --user daemon-reload
# You can now start/enable each remote by using rclone@<remote>
#   systemctl --user enable --now rclone@dropbox

[Unit]
Description=rclone: Remote FUSE filesystem for cloud storage config %i
Documentation=man:rclone(1)
After=network-online.target
Wants=network-online.target

[Service]
Type=notify
ExecStartPre=-/usr/bin/mkdir -p %h/mnt/%i
ExecStart= \
  /usr/bin/rclone mount \
    --config=%h/.config/rclone/rclone.conf \
    --vfs-cache-mode writes \
    --vfs-cache-max-size 100M \
    --log-level INFO \
    --log-file /tmp/rclone-%i.log \
    --umask 022 \
    --allow-other \
    %i: %h/mnt/%i
ExecStop=/bin/fusermount -u %h/mnt/%i

[Install]
WantedBy=default.target

Work nicely form me for first run, but after restart, I get error nonempty directory:
024/01/15 21:35:36 mount helper error: fusermount: mountpoint is not empty
2024/01/15 21:35:36 mount helper error: fusermount: if you are sure this is safe, use the 'nonempty' mount option
2024/01/15 21:35:36 Fatal error: failed to mount FUSE fs: fusermount: exit status 1

Trying to use option nonempty, but get:
2024/01/15 21:36:50 ERROR : -o/--option not supported with this FUSE backend

Running Proxmox 8. Help appreciated. Thanks a lot

Sure! Here is a better version of the text:

Hello, I'm having some trouble with the drive mounting script. When I mount the folders, they appear empty in Nautilus. I'm using two Google Drive accounts, A and B, each with two namespaces: personal and shared. The file structure is as follows:

~/drive/{account}/{namespace}

This works just fine using Rclone Browser. However, when I use the script provided above, I can mount the folders to /mnt/drive/%i, but they are empty.

@yanisallouch Did you make the change to /etc/fuse.conf mentioned in the preparation section? It is required for many desktop environments and will result in the "empty folder" issue you're seeing if you skip it.

@kabili207 Yes :

# The file /etc/fuse.conf allows for the following parameters:
#
# user_allow_other - Using the allow_other mount option works fine as root, in
# order to have it work as user you need user_allow_other in /etc/fuse.conf as
# well. (This option allows users to use the allow_other option.) You need
# allow_other if you want users other than the owner to access a mounted fuse.
# This option must appear on a line by itself. There is no value, just the
# presence of the option.

user_allow_other


# mount_max = n - this option sets the maximum number of mounts.
# Currently (2014) it must be typed exactly as shown
# (with a single space before and after the equals sign).

#mount_max = 1000

Thanks @kabili207 and everyone else who contributed here! Very useful indeed.

I created a small repo out of all this - if you have any objections, please let me know and I'll take it down or modify it accordingly.

Hi, my service stopped working all of a sudden. It would fail to start after reboot and reloading daemon. I reinstalled fuse3 and Plasma 6 , and removed the --allow_users flag. After this the service started fine. I realized that something might have broken with updating this week(Arch on LTS kernel). When I checked the /etc/fuse.conf file user_allow_other was commented. This was un-commented before. I also tried it with the --allow_users flag again and user_allow_other un-commented and it worked as before. Just an FYI.

Just wanted to say thanks for this - works great for me

Hi, my service stopped working all of a sudden. It would fail to start after reboot and reloading daemon. I reinstalled fuse3 and Plasma 6 , and removed the --allow_users flag. After this the service started fine. I realized that something might have broken with updating this week(Arch on LTS kernel). When I checked the /etc/fuse.conf file user_allow_other was commented. This was un-commented before. I also tried it with the --allow_users flag again and user_allow_other un-commented and it worked as before. Just an FYI.

Good point, thanks. /etc/fuse.conf gets overwritten on upgrading whenever it's content in the package changes, and multiple config file loading like /etc/apt/sources.list.d/* isn't yet supported AFAIK.

If you don't need to access the mount with users other than your own, it's actually safer and easier to just remove --allow-other from rclone mount options, which doesn't require setting user_allow_other.

Tested on Debian 12.5, rclone 1.66

gave this a go but no luck, I get this Job for [email protected] failed because the control process exited with error code. See "systemctl --user status [email protected]" and "journalctl --user -xeu [email protected]" for details.

@Nathan22211 I would recommend running the two commands it suggests to see what went wrong

ok well it's definitely something with fuse... Failed to start rclone: Remote FUSE filesyste m for cloud storage config gDrive. I have fuse 3 and fuse 2 on my system. fuse3 for rclone and fuse2 for orca slicer

That's just systemd taking the description from line 11 and saying it couldn't start it.

Try copying the command on lines 20 through 28 and running it directly in the terminal. You'll need to replace the values %h with the path to your home directory and %i with the name of the remote when you do this

yeah that did nothing at all

This script works well, thanks for your contribution, @kabili207. Maybe this is a silly question, but I can't seem to get it right. How do I mount a drive to a remote location's subdirectory?

I've tried the following where DropBox is the rclone remote name Photos/0test is the subdirectory I'd like to mount:

• systemctl --user enable --now rclone@"DropBox:Photos/0test/"
• systemctl --user enable --now rclone@DropBox:Photos/0test/
• systemctl --user enable --now rclone@DropBox Photos/0test

BTW, when I run systemctl --user enable --now rclone@DropBox it does in fact mount my root DropBox folder and works as expected.

I've also looked into rclone config to see if I can statically enter the remote subfolder in the config file but I haven't figured that out either.

Any help would be greatly appreciated.
Cheers.

@jhub95 I can't say I've ever tested using a subdirectory, but looking at the docs for the rclone mount command, you'll probably want to edit line 28 of the service file to remove the : so you can specify it manually.

Keep in mind that if you do this you'll likely have to update anything that's not using a subdirectory so the remote ends with a : like so systemctl --user enable rclone@remote:

In my experience, the -z option (lazy unmount) can be useful in some cases for safely unmounting rclone mounts; at least, its use is not harmful.

Another suggestion: When using systemd, it doesn't seem appropriate to write log output directly to a file. Why not use systemd itself for this?

Instead of:

--log-level INFO \
--log-file /tmp/rclone-%i.log \

Suggestion:

--log-level INFO \
--log-systemd \

...Rolf

I'm glad you caught that because I added the logging for testing initially and completely forgot about it

Hi, and thanks for your template. I have two things want to mention though.
The rclone mount#systemd document says:

you should provide --config and --cache-dir explicitly as absolute paths via rclone arguments...

But this template only provides the config path.
Second, the systemd specifiers are already provided XDG_CONFIG_HOME(as %E) and XDG_CACHE_HOME(as %C), so I guess I can use them directly(?)
I use this for my computer:

ExecStart=/bin/rclone mount \
    --config="%E/rclone/rclone.conf" \
    --cache-dir="%C/rclone" \
    --vfs-cache-mode=writes \
    "%i:" "%h/mnt/%i"

By the way, I'm on archlinux with kde plasma, and I don't know why, but I didn't need the --allow-other. Just saying.

@rolfn

In my experience, the -z option (lazy unmount) can be useful in some cases for safely unmounting rclone mounts; at least, its use is not harmful.

I didn't even know about "lazy unmount" this concept in the past. This should solve the problem caused by incomplete data transfer during unmounting. I've already lost some test data before.

Suggestion:

--log-level INFO \
--log-systemd \

They didn't document this flag on the website, almost missed it.

Here I had to give a few more commands to get it to run smoothly.

mkdir -p ~/.config/systemd/user/

vim ~/.config/systemd/user/[email protected]
# PASTE SERVICE FILE AND SAVE

loginctl enable-linger $USER
export XDG_RUNTIME_DIR=/run/user/$(id -u)
systemctl --user daemon-reload
# rclone's remote called gd in ~/.config/rclone/rclone.conf
systemctl --user enable --now [email protected]

# show the files
ls -lah ~/mnt/gd

Thank you sir.