I noticed today that two of my containers were down. Tried to fire them back up and received a very generic "execution error" "server error" message. Took a look at Settings > Docker and found a message stating "one or more paths to not exist" (only shown in basic view). So I went to the CLI I found this, which can't be good:

drwxrwxrwx  1 nobody users 272 Feb 14 15:28 cache/
drwxrwxrwx 15 nobody users 293 Jan 31 04:30 disk1/
drwxrwxrwx  9 nobody users 149 Nov 17 16:37 disk2/
drwxrwxrwx  9 nobody users 158 Feb 14 15:28 disk3/
drwxrwxrwt  3 nobody users  60 Nov 15 10:33 disks/
drwxrwxrwt  2 nobody users  40 Dec  9 09:00 remotes/
d?????????  ? ?      ?       ?            ? user/
drwxrwxrwx  1 nobody users 293 Feb 14 15:28 user0/

Looks like the user path has gone away somehow. I am just looking through diagnostics now, but I don't do that often. If someone is faster and has time, I would appreciate any advice in knowing what happened and what I should do to correct. I am tempted to just reboot, but will be patient since I am worried that might put me hard down situation. Some things like VMs are still functioning fine, I suppose they reference the cache path, rather than user.

 

Diags attached.

 

Thanks!

-Landon

anonymized-diagnostics-20210214-1549.zip

Well, I couldn't see anything obvious in the logs as to a cause. So I went ahead and rebooted. Everything came back up fine, but it does make me uneasy. If anyone has any wise ideas on what might have happened I would be excited to hear it.

Nothing in the environment has changed for months that I can think of. The only thing that ran over night last night when this seemed to occur were some SSH backup sessions where the server is the target and also an SSD TRIM. I don't see much else...

 

Thanks!

It's a known issue:

 

https://forums.unraid.net/bug-reports/stable-releases/683-shfs-error-results-in-lost-mntuser-r939/

Thank you for pointing me to that. I guess in my panic I didn't search very well.