Server hangs when stopping array on a semi-regular basis

[romain]

I tried stopping a docker container that was unresponsive this morning and was getting an error so I tried to stop the array and now it looks like it's having trouble unmounting a zfs cache pool. I ended up restarting the server manually.

 

These lines are repeating in the log:

 

Jan  8 12:07:12 unRAID root: rmdir: failed to remove '/mnt/user': Device or resource busy
Jan  8 12:07:12 unRAID emhttpd: shcmd (6600): exit status: 1
Jan  8 12:07:12 unRAID emhttpd: shcmd (6602): rm -f /boot/config/plugins/dynamix/mover.cron
Jan  8 12:07:12 unRAID emhttpd: shcmd (6603): /usr/local/sbin/update_cron
Jan  8 12:07:12 unRAID emhttpd: Retry unmounting user share(s)...
Jan  8 12:07:17 unRAID emhttpd: shcmd (6604): /usr/sbin/zfs unmount -a
Jan  8 12:07:17 unRAID root: cannot open 'trash': pool I/O is currently suspended
Jan  8 12:07:17 unRAID root: cannot open 'trash/media': pool I/O is currently suspended
Jan  8 12:07:17 unRAID root: cannot unmount '/mnt/cache': pool or dataset is busy
Jan  8 12:07:17 unRAID emhttpd: shcmd (6604): exit status: 1
Jan  8 12:07:17 unRAID emhttpd: shcmd (6605): umount /mnt/user
Jan  8 12:07:17 unRAID root: umount: /mnt/user: target is busy.
Jan  8 12:07:17 unRAID emhttpd: shcmd (6605): exit status: 32
Jan  8 12:07:17 unRAID emhttpd: shcmd (6606): rmdir /mnt/user

 

This happens on and off, I've googled and searched the forum before for answers but am not sure what is going on.

 

I ran diagnostics before forcing the reboot, they're attached here. Hopefully someone can point me in the right direction to fix whatever is causing this.

 

unraid-diagnostics-20240108-1152.zip

[JorgeB]

Jan  8 11:25:14 unRAID emhttpd: shcmd (3652): umount /var/lib/docker
Jan  8 11:25:14 unRAID root: umount: /var/lib/docker: target is busy.

 

Problem starts with docker, may be a specific container, you can also try 6.12.7 which should be out soon, there are some changes to try and handle that better.

[romain]

Thanks @JorgeB I will wait. This has been happening since late 6.11.x releases but I've also reconfigured a pool and added a pool so wasn't sure what was going on.