April 5, 20251 yr Hello everyone, I'm currently setting multiple cache/pool and I think I did something per mistake and I'm looking at the info. Right now, I'm moving most of the stuff that were on my "cache" to a new pool. Thing is when I moved the file from the cache to the pool, it got from 352gb to nearly 700gb. Weird thing is I don't even have move my appdata yet which is the 300gb. This is what I have set to be on the cache pool: - Appdata - domains - lxc - mcron - syslog - system I have already moved everything except appdata. Appdata is on the array currently (dockers, vm, lxc are all shutdown and disabled). System is the one that's about 650gb by itself. I'm wondering if system was on the cache or not but looking at the docker configuration, it says /mnt/user/system/docker should be on the ssd so I'm pretty sure it was. But is it possible not everything from /mnt/user/system was on the cache? Or maybe before it was /mnt/cache/system and I changed it somewhere in the past? Thus this post is to get the default value of these share normally or the configuration so I can find out where's the problem thank you edit: I'm currently rebuilding my parity (error which invalidated my parity disk) and I see that the folder /mnt/user/system/docker/btrfs is gaining in size non stop. It seems the 600gb+ is coming from there. But docker is currently disabled... Edited April 5, 20251 yr by Nodiaque
April 5, 20251 yr Community Expert Nothing can move open files, so you have to go into Settings and Disable Docker and VM Manager before appdata, domains, system shares can be moved. No good reason for system share to be that large, so if it is, it is for a bad reason. Attach Diagnostics ZIP to your NEXT post in this thread.
April 5, 20251 yr Author Everything was disabled before moving (vm, lxc, vm). I'm wondering if I corrupted my docker subsystem. Haven't started the vm yet, I'm waiting for the parity to be restored first. I'm running the diagnostic right now but is it normal that it's scanning these files? Like my whole appdata folder. edit: what I don't get is that /mnt/cache is empty so what is it scanning. Appdata is currently under /mnt/user/appdata on the array, not on the cache. Edited April 5, 20251 yr by Nodiaque
April 5, 20251 yr Author It's been scanning my jellyfin folder (well, it says it's scanning /mnt/cache/appdata but that folder is empty) for the past 30 minutes and still doing it. I don't think it's right.... edit: an hour later, still no diagnostic yet, it's still scanning appdata.... Edited April 5, 20251 yr by Nodiaque
April 5, 20251 yr Community Expert Do you have a disk4? What do you get from command line with this? du -h -d 1 /var/log
April 5, 20251 yr Author No disk 4 was an error during initial allocation. I kept it because I'll grow the array one day so the slot is ready.
April 5, 20251 yr Community Expert Go to Tools - System Log, DOWNLOAD and post the zip. Then reboot and see if you can get diagnostics.
April 5, 20251 yr Author I tried to do that, after more then an hour of looking at the screen doing sed -i /mnt/cache/appdata (a folder that doesn't exist at that location), the tab crash and I never got the zip file. You can see the screenshot of these output in a reply higher. Right now, I'm rebuilding parity, 9h left. I'll reboot after. edit: I tried stopping the parity sync and running the diagnostic, it still try to get sed from that /mnt/cache/appdata path Edited April 5, 20251 yr by Nodiaque
April 5, 20251 yr Community Expert The command line results screenshots you posted are clipped. See if you can get this one again and post the entire result. du -h Also 4 minutes ago, trurl said: Go to Tools - System Log, DOWNLOAD and post the zip.
April 5, 20251 yr Community Expert The reason I am asking for syslog now is because it will be reset after reboot and it might tell us something if we can get it before that.
April 5, 20251 yr Author 1 minute ago, trurl said: The reason I am asking for syslog now is because it will be reset after reboot and it might tell us something if we can get it before that. Oh my syslog is saved on my share so I never lose them.
April 5, 20251 yr Community Expert Your syslog is filling with mover entries. And it seems to be moving appdata, which usually shouldn't be moved (unless it is on the array where it shouldn't be). Disable mover logging in Settings - Scheduler - Mover Settings. We can enable it again later if we need to diagnose that further.
April 5, 20251 yr Author the mover was before the parity check. I had mover running this morning but right now, it's not working. I enabled mover logging specially to see what it was doing. If you scroll down the log, you'll see the mover was about 8 hours ago and the last entry are the pause and resume of parity check
April 5, 20251 yr Author Just now, trurl said: Are you using the Mover Tuning plugin? yes Current log: No mover since at least 14h00
April 5, 20251 yr Community Expert Looks like it was also moving lxc. I don't use that but I would assume it has a purpose similar to system and/or appdata shares, so it probably shouldn't be moved either. We'll want to uninstall Mover Tuning to eliminate any unusual behavior that might cause. We can wait until parity has finished rebuilding to continue diagnosing.
April 5, 20251 yr Author Tuner to be honest, I don't really use it. All I have is to have a schedule to run it, everything else is default. The reason you see lxc and stuff is because I'm moving from one ssd pool to another one. I though about using mover for that so I put Cache => Array in my share config and ran mover while everything was disabled (lxc, vm, docker). When it ended, I moved the system folder manually and after about 6 hours, that's when I saw it had grew to about 800gb which made no sens. I'm planning on moving back everything to the original pool from the array once parity is done and check if stuff still work.
April 5, 20251 yr Community Expert 1 minute ago, Nodiaque said: Tuner to be honest, I don't really use it. All I have is to have a schedule to run it, everything else is default. Built-in mover already has scheduler so Mover Tuning not needed. Unraid V7 can move between pools, no need to go to array. https://docs.unraid.net/unraid-os/release-notes/7.0.0/#share-secondary-storage-may-be-assigned-to-a-pool
April 5, 20251 yr Author yeah, I installed it in the past thinking I could have what I wanted out of it. Turns out no and I never uninstall it. Unraid 7 is my next move to have that new between pool feature since my current setup require that Edited April 5, 20251 yr by Nodiaque
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.