DZMM

That's fine. What's supposed to happen is the /mnt/user/appdata/other/rclone/rclone_mount_running file gets removed when you run the unmount script either at shutdown or array start. Just manually delete it for now: /mnt/user/appdata/other/rclone/rclone_mount_running I designed the mount script this way so you can run it on a cron job to auto remount if the mount ever fails. This check makes sure that 2 instances of the script don't run at the same time

This thread has got a lot more action than I or @Kaizac or @slimshizn probably ever hoped for. There's been a lot of action getting people up and running, so I'm wondering how some of the people in the background are getting on? How intensively are other people using rclone? Have you moved all your media? How are you finding the playback experience? Personally, I don't have any plex content on my unRAID server anymore, except for my photos and I've now got over 300TB of Plex content stored on gdrive, as well as another big crypt with my backups (personal files, VM images etc). I don't even notice the impact of streaming anymore and when I do have any skips, I actually think they are because of local wi-fi issues rather than rclone.

fixed - my mistake - an error sneaked in on github https://github.com/BinsonBuzz/unraid_rclone_mount/blob/master/rclone_mount. you need to add a new section at the start of your mount script - ####### Check if script is already running ########## if [[ -f "/mnt/user/appdata/other/rclone/rclone_mount_running" ]]; then echo "$(date "+%d.%m.%Y %T") INFO: Exiting script already running." exit else touch /mnt/user/appdata/other/rclone/rclone_mount_running fi ####### End Check if script already running ########## your first run was actually successful. What failed was the docker start command as you don't have those dockers or they have slightly different names on your server. Remove or edit I'm not really sure what happened on the 2nd run...I think your first mount is lingering somewhere. Manually run: fusermount -uz /mnt/user/mount_unionfs/google_vfs fusermount -uz /mnt/user/mount_rclone/google_vfs before running the mount script again. Once you've edited the docker lines and added the missing section to the mount script you should be good to go as long as you've used the latest upload script from github

you need to install nerdtools to then install unionfs - see the first post for the 'optional plugins' section post your full mount script as this is easily fixed

Everybody's nice on this forum! Can you post your rclone config file (remember to remove any passwords), mount script and the log with any errors in the thread please. It's easier to follow and I couldn't open your links anyway, although it could be because I'm on my work laptop which is seriously locked down.

you've got something in /mount_unionfs/google_vfs/ prior to the unionfs mount being created so it fails. Manually delete whatever's in there (be sure you don't want first!)

Neither do I really! I cobbled these scripts together using other guides and Google.

oh yeah, Krusader seems to have problems if you don't restart it. I don't use so I'm not sure why - I use mc to move files

All looks ok to me - maybe someone else will spot something @Kaizac ??. What happens if you create a few test folders in: (i) /mnt/user/mount_unionfs/google_vfs (ii) directly in /mnt/user/rclone_upload/google_vfs Do they appear in /mnt/user/mount_rclone/google_vfs? (change --min-age 30m to --min-age 1m or delete in the upload script to test) The only thing I can think of is there was a weird gdrive issue last week where uploads from Europe were blocked for a day, but this has been resolved as far as I know

Can you post your rclone config and scripts please

16gb should be enough. I'd disable the scripts and reboot. Then delete any residual checker files form /mnt/user/appdata/rclone (i think that's where all the script checks are) and then enable the scripts.

your unionfs mount is failing, I think because you created your movies and tv_shows folder incorrectly. you need to delete those folders and then run your mount command/script again, then recreate the folders and they'll be copied to gdrive.

how much ram do you have? Does rebooting help?

It does it when it can't see the files ie the mount isn't mounted. If you want fix ahead of schedule tell Plex to do a manual scan Is all working now?

Does /mnt/user/mount_rclone/google_vfs exist? The Plex behaviour is correct as rclone isn't mounted, so it can't see the files. It will self correct once you fix the mount.

Thanks for posting the thread. I hadn't noticed any problems with playback and none of my users had mentioned anything, but my upload job was getting 429 errors. Adding the user agent to my upload script seems to have got it going again but it seems a bit slower - about 75% of the previous speed. It's a bit worrying if they are targetting the rclone user agent.....

Manually remove this file /mnt/user/appdata/other/rclone/rclone_upload It should get removed at array start/end depending on how you've setup your unmount script

You either need to put the files in the right subfolders in your upload folder, or your scripts aren't quite setup properly - post your upload script

The scripts don't move files to the upload folder - you have to do that, or you've setup radarr to add new files to the mount. The min-age setting tells rclone not to move files from the upload folder to the mount that's younger than 30 mins.

The touch command creates a file - you need to run the next commands to copy it to the mounts: rclone copy mountcheck gdrive_media_vfs: -vv --no-traverse rclone copy mountcheck tdrive_media_vfs: -vv --no-traverse

Playback will be awful if you stream using a normal mount and you'll probably suffer from API bans

Can you post your rclone config (remember to take out any personal info e.g. passwords) and mount script please

not sure what's happening, but try taking: mkdir -p /mnt/user/mount_rclone/google_vfs out of your mount script

Just configure tdrive to not be a crypt and make it a direct mount of your google drive.

Yes, although I'd do every 5 minutes personally and also run the unmount at array start so it runs if you have an unclean shutdown