My apologies, I have tried to sift through the many posts regarding this error, but I haven't been able to find a solution that works for my server.

Context: when I restart the server, initially my docker services start and run correctly, but after a few minutes the services stop functioning correctly.

If I stop the docker service and then restart it, I get the "Docker Service failed to start" error.

I am not able to delete any of the docker service images, nor can I delete the docker.img file to try and rebuild the image.

I have also tried increasing the size the vDisk size of the image to 60GB with no change.

I have uploaded the syslog.txt file (and diagnostics files)

Any help would be greatly appreciated.

syslog.txt

tower-diagnostics-20260408-2159.zip

Edited April 9Apr 9 by bob_loblaw

Attach Diagnostics ZIP to your NEXT post in this thread.

Just some advice on using the forum. When you edit a previous post, anybody that has already read the post, and might be willing to help (such as me) will not know there is anything new to see in the thread, and so may never visit the thread again unless someone actually does make a new post in the thread. That is why I said

47 minutes ago, trurl said:

Attach Diagnostics ZIP to your NEXT post in this thread.

Before we actually try to fix your dockers, let's get your appdata off the array.

Ideally, Docker/VM related shares - appdata, domains, system - should have all files on cache or other pool, with nothing on the array, so Docker/VM will perform better, and so array disks can spin down since these files are always open.

domains and system already have no files on the array, but appdata does have files on the array.

Mover won't move those files because appdata has no Mover action, because it has no Secondary storage.

Set appdata to Primary:cache; Secondary:array; Mover action:array->cache.

Nothing can move open files. Go to Settings and disable Docker.

Run Mover, wait for it to complete, post new diagnostics.

Thanks for the advice. I will make sure to do that next time.

Thanks - Appdata, Domains and System folders are all moved to cache.

New diagnostics file attached.

tower-diagnostics-20260409-0010.zip

Docker troubleshooting | Unraid Docs

This page offers troubleshooting guidance specifically for Docker containers on Unraid, managed by the Unraid team. For advanced troubleshooting, in-depth technical details, or topics beyond the Unrai

I have disabled Docker and attempted to delete the vDisk file (docker.img) by clicking the checkbox and then clicking delete at the bottom, but when I do the page just returns with the "Delete vDisk file" option still available (which I understand is a sign it didn't work.

I have also tired to delete the file directly through the file manager, but it will not delete.

Also BTW, each time I increased the max size of the vDisk file in the Docker settings, the size of the actual docker image increased as well.

Have you done that since those last diagnostics? Might be useful to see syslog that includes those activities.

Here is the most recent diagnostics file. Thanks

tower-diagnostics-20260409-0244.zip

Apr 9 00:03:48 Tower kernel: BTRFS error (device nvme0n1p1 state A): Transaction aborted (error -5)

Apr 9 00:03:48 Tower kernel: BTRFS: error (device nvme0n1p1 state A) in __btrfs_free_extent:3234: errno=-5 IO failure

Apr 9 00:03:48 Tower kernel: BTRFS info (device nvme0n1p1 state EA): forced readonly

Pool is going read-only, and since BTRFS detected a few corruption errors, my first recommendation is to run memtest.

Apr 8 23:57:28 Tower kernel: BTRFS info (device nvme0n1p1): bdev /dev/nvme0n1p1 errs: wr 0, rd 0, flush 0, corrupt 8, gen 0

If nothing is found for now, backup and recreate the pool and keep monitoring. If new corruption errors show up, there's still a problem.

Apologies for the delay in following up.

I've run memtest, and it identified a stick of ram that had errors, so I removed that stick and retested, and found no further ram errors.

That did not solve the issues, so I updated the bios on my motherboard to the latest, but again the errors persist.

I have tried to erase the plugins and the vdisk files, but it doesn't look like anything happens when I do so.

I've attached the latest diagnostics files to this post.

Thanks

tower-diagnostics-20260421-2057.zip

Fixing the RAM issue won't resolve the existing damage done to the pool filesystem; the pool is still corrupt. Recommend backing up what you can and reformatting.

Thanks JorgeB, and please forgive my newness here, but are there any potential "fixes" I can try to employ before embarking on a 28TB backup and reformatting?

The problem is with the cache pool only for now; no need to backup and reformat the array.