nerv

I moved my docker folder back to my cache disk and that appears to have fixed the issue as everything is much faster. Not surprising but frustrating.

Hmm okay, it's on a 120TB pool though, but I was getting alerts before about docker space available that corresponded to the disk not the pool/share. Mostly I just want to make sure of the size before I move it back to my cache to see if it resolves my issues with updating taking so long they timeout. I guess I just need to wait for du -sh to finish to see how big it is?

I expect this issue predates 7.0. I noticed unraid is reporting my docker folder size as 3.89 TiB. Looking through the directories I traced it to overlay2, but it's so big with so many files du -sh just times out. Setting max depth 1 I can see there's some multi gig folders in here (just at 1 level). I've run docker system prune --all etc and nothing is getting cleaned up, and docker system df is showing ~15 GB of space. Any ideas why this folder is so big or advice on how to debug further? I'm running about 25 dockers, fairly standard plex, *arr, etc.

on 7.0, same issue

Did this ever get solved? I have the same problem. Is the timeout update the recommended solution?

FYI I did move the docker folder to another disk with higher capacity and the error went away. So that pretty much confirms the check is looking at the disk's space rather than the share it belongs to. Appreciate the help!

got it. I'm a little terrified to move the folder 😂, but agreed that's a good way to confirm the issue.

Sorry not fully following, are you saying the error message is correct? I do have split any directory enabled, so the disk capacity should not matter.

Looking at this more carefully, I think it's because the docker folder is on disk 4, which is at 13.1/14TB that rounds up to, 94%. That disk has been roughly at that usage since I changed to the docker directory a few months ago, because the array has been filling other lower disks the entire time. However, given the share is set to split at any directory, and can use all disks, perhaps the error is wrong? If the directory couldn't be split to another drive, or it was only allowed on that Drive it would make sense. But in practice won't unraid split this directory to another disk at some point?

Some of this just dates back to what was available at the time. E.g. I had some disks on reiser until a couple years ago until I took the time to migrate them all. I've been on Unraid for over a decade by now. Last I checked XFS was recommended for the array and BTRFS was what was available for the pool 🤷‍♂️ I may move the docker image back to the cache, but I haven't had performance issues as is. If there's nothing else to try I'll just silence the error. Fwiw I'm skeptical it has anything to do with free space on disks etc. The warning is complaining about the image (I don't have one) and 94% was the warning I got before migrating to docker directory to fix that error, and it has consistently been warning 94% for ~3 months despite the array changing in size a bunch since then (though admittedly most of the disks will have remained the same usage in that time). Seems like a stale error that won't go away but it doesn't really matter, I can ignore it. Appreciate the input.

No, the share is set to store on the Array.

What does the plugin base Docker image file is getting full on when using a docker directory? My docker image was getting full but I switched to a directory months ago. I deleted the old image files, and have rebooted the system, but I'm still getting this error. Afaik I have no docker image, so I don't know how to make this error go away. I can ignore it, but I'd rather not. thanks!

Closing this out, after 2 weeks no crashes. Seems very likely the UPS was just failing.