I needed to make a few drive changes to my primary server. I stopped all Dockers, unmounted all NFS connections from other machines, and hit the button to stop the array. It kept throwing an error about not being able to unmount the /mnt/user folder. With the Open Files plugin, I found that there was a process running that was preventing the unmount of the array, SHFS. I dropped to a command line, sent a 'kill -9' command to kill that process, and the array finished shutting down. I then did a new array config, plus added a 3-drive zpool. Clicked the button to start the array and the log was filling with syslog errors. Digging around, I found that all of my shares are missing, as well as the '/mnt/user' folder in the root.

neo-diagnostics-20241223-0539.zip

Ok, I don't know WTF is going on with my server. I decided to reboot it 'just to try' and the shares were back, but the 'appdata' was attached to a non-existent location. I had to reconfigure it, but it's back in the proper location now, and appears to be intact. In fact, it seems to be completely fine now.

Something went absolutely stupid when I stopped the array and made the changes. Hopefully the diagnostics point to WTH happened.

Dec 23 05:29:54 Neo emhttpd: shcmd (2814289): mkdir /mnt/user
Dec 23 05:29:54 Neo root: mkdir: cannot create directory '/mnt/user': File exists

 

This was what caused the issue.

36 minutes ago, JorgeB said:

Dec 23 05:29:54 Neo emhttpd: shcmd (2814289): mkdir /mnt/user
Dec 23 05:29:54 Neo root: mkdir: cannot create directory '/mnt/user': File exists

 

This was what caused the issue.

But, it didn't exist. When I went in the File Browser, it wasn't there. Nor was it in the /mnt directory from the command line. It wasn't there until I did the reboot.

For mkdir to report that, I think it must have existed, but a reboot will always fix this, since it's created during array start.