DZMM

yep - that's what I thought as well. Options I choose are: - 2 / Encrypt the filenames - 1 / Encrypt directory names - 1 Password or pass phrase for encryption. y) Yes type in my own password - Password or pass phrase for salt. y) Yes type in my own password

I use the same passwords for both mounts, different IDs (tried same didn't fix the problem) but different tokens because using two different google accounts. I'm not sure what rclone uses to encrypt but I'm guessing the tokens are the reason or the actual remote name. I'm hoping there is a fix. But, it's not a big problem as everything 'works' - it'd just be nice to have the option of being able to move files between remotes. It's made such a big difference today having another another 750GB or 70Mbps upload, as the single 750GB limit has been a pain in the ass for the last couple of months as I've had to manually manage how fast my rclone_upload backlog grew whereas now it's finally shrinking. I think I'm going to create a 3rd team drive tomorrow.

unfortunately I can't move files as the encryption is different - I think because I'm using different tokens. I've asked on the rclone forums if there's a way around this because I'd like to periodically move all files to one location

I just made a very useful change to my scripts that has solved my problem with the limit of only being able to upload 750GB/day, which was creating bottlenecks on my local server as I couldn't upload fast enough to keep up with new pending content. I've added a Teamdrive remote to my setup, that allows me to upload another 750GB/day in addition to the 750GB/day to my existing remote. This is because the 750GB/day limit is per account - by sharing the teamdrive created by my google apps account with another google account I can upload more. Theoretically I could repeat for n extra accounts (each one would need a separate token team drive), but 1 is enough for me. Steps: create new team drive with main google apps account share with 2nd google account create new team drive remotes (see first post) - remember to get token from account in #2 not account in #1 otherwise you won't get 2nd upload quota amend mount script (see first post) to mount new tdrive and change unionfs mount from 2-way union to 3-way including tdrive new upload script to upload to tdrive - my first upload script moves files from the array, and the 2nd from the cache. Another way to 'load-balance' the uploads could be to run one script against disks 1-3 and the other against 4-x add tdrive line to cleanup script add tdrive line to unmount script Optional repeat if need more upload capacity e.g. change 3-way union to 4-way

if you set your appdata, mount_rclone and mount_unionfs to cache only that should do the trick. I thought touch created the necessary directory tree? If not, glad you figured out how to fix.

That's a lot - check out https://www.hetzner.com/sb or https://github.com/Admin9705/PlexGuide.com-The-Awesome-Plex-Server/wiki/Recommend-Hosting-Servers @nuhll that's a bad way to do it IMO. How much are you spending? You could probably spend the same on a dedicated server from the links above and get gigabit speeds and processing power to run your plex server without using any of your home bandwidth

What's your VPS spec and how did you set it up? I might have to do this next year as I might be moving from 300/300 to an area where the upload is only about 30. Going to read up on 2.4.4_1 now

I wouldn't rely on resume working with gdrive as it doesn't work yet https://github.com/ncw/rclone/issues/87

Safest way is to create a new remote gdrive_new: and add this to your scripts IMO the script sounds like a bad idea. Upload doesn't resume so if you've spent 2 hours to reach 99% of a file uploaded and then stop, you'll have to start all over again. I'd recommend using bwlimit to schedule peak and off-peak upload speeds, or just do overnight. If you have to police your upload i.e. you can't comfortably upload 500-750GB/day then the service becomes less useful. Edit: or find a way to traffic shape the rclone upload e.g. pfsense

@nuhll is everything else working ok now?

Anyone else having problems with radarr & sonarr connecting to Jackett? Edit: ignore - somehow my API key changed

yes, the mountcheck file is created outside of rclone by running these two commands which creates the file and then moves it to the root of the rclone mount: touch mountcheck rclone copy mountcheck gdrive_media_vfs: -vv --no-traverse once done you should have something looking like this when looking locally at mount_rclone and mount_unionfs: and on drive the mountcheck file is encrypted so it will have a funny name. I went back through this thread and your rclone remote names are the same as mine, so running the two commands above in terminal will definitely do the job.

It's earmarked for the next release, so fingers crossed https://github.com/ncw/rclone/milestone/33

If you are renaming files you've already uploaded, unfortunately this is what unionfs does. Hopefully rclone union in a future release will allow unraid users to ditch unionfs

Hi I'm having another go at getting this working. I have LL adding books to my calibre library folder and each book gets a title (#) entry so it looks like all is correct. However, Calibre doesn't see the new books until I restart it - is Calibre supposed to be in sync with LL?

What I find is that if you move files in the mount by moving them in the unionfs mount, unionfs downloads the file and reuploads it. What I do if I'm doing a big move like that, is I move the files manually within the rclone mount as they don't get downloaded then

It's saying rclone isn't installed - check plugins

it doesn't matter as no real files exist on your array for those folders - just virtual files. Just add the current folders to your unionfs mount and then move them manually behind the scenes to your rclone_upload folder where they'll get uploaded over time e.g. unionfs -o cow,allow_other,direct_io,auto_cache,sync_read /mnt/user/rclone_upload/google_vfs=RW:/mnt/user/mount_rclone/google_vfs=RO:/mnt/user/OLD_MEDIA_LOCATION=RO /mnt/user/mount_unionfs/google_vfs Then add the unionfs folders to sonarr etc. You just have to make sure that the filepaths matchup e.g. - /mnt/user/mount_rclone/google_vfs/TV Show 1 - /mnt/user/Old/Media_location/TV Show 1

rc doesn't like it if you keep unmounting and mounting - it tends to only work the first mount. Don't worry about the errors that's fine. The purpose of the rc command is to pre-populate the directory cache at mount to ensure navigating the mount is as fast as possible from the outset. The dir cache stills gets populated without it, just not as fast.

Thanks - script picked up errors on my UD pool which I've corrected

What scrub arguments do you use for a UD pool? Thanks

I just followed @Symon earlier post blindly All good now, will set 19218 to Node 1 strict later: Per-node process memory usage (in MBs) PID Node 0 Node 1 Total ----------------------- --------------- --------------- --------------- 10236 (qemu-system-x86) 8269.48 0.00 8269.48 19218 (qemu-system-x86) 4146.88 181.60 4328.48 124935 (qemu-system-x86) 19.52 8257.12 8276.64 127416 (qemu-system-x86) 8264.96 0.00 8264.97 ----------------------- --------------- --------------- --------------- Total 20700.85 8438.73 29139.58

I've just done 3 of my 4 VMs and my memory usage looks the same for interleaved: Per-node process memory usage (in MBs) PID Node 0 Node 1 Total ----------------------- --------------- --------------- --------------- 18388 (qemu-system-x86) 4143.74 167.77 4311.51 29504 (qemu-system-x86) 4864.92 3407.02 8271.94 36634 (qemu-system-x86) 4881.64 3376.45 8258.08 115105 (qemu-system-x86) 8236.66 25.16 8261.83 ----------------------- --------------- --------------- --------------- Total 22126.96 6976.40 29103.36 18388 is the VM I haven't tweaked yet as it's my pfsense firewall, so I have to do that when the family aren't doing stuff. Unfortunately all the memory is on Node 0, when the cores are on 1. I'm wondering if memory mode isn't working because the memory is already in use e.g. by unRAID, dockers etc so if there's not enough when the VM is created unRAID uses memory from the other node? Otherwise to allocate say 8GB when there's only 6GB available it would have to move other blocks around????