I've rebooted my server twice in the last week or two. First time, I hit the "Reboot" button under Array Operations on the Main tab, the second time, I clicked the "Reboot now" button after installing the 7.3.1 upgrade (from 7.3.0). I have no idea what might have caused these unclean shutdowns, since I asked the server to do it, and it wasn't the result of a power outage or other unexpected situation. This afternoon, I tried shutting down the VM service, then rebooted, and it came up clean. I don't know if that had anything to do with it or not.

Diagnostics are from after this most recent (successful/clean) reboot, though I don't know if they'll be all that helpful.

I'm also attaching the copy of the syslog that's being written to a pool by the Syslog Server (don't know if that's got any older info that may not be included in the diagnostics). The upgrade & reboot happened on 1 June with the reboot starting at "Jun 1 16:54:29" (to make searching easier).

If anyone can find a reason that the shutdown was unclean, I'll at least have an idea of what to watch for before doing a restart in the future. As it stands, it seems that shutting down the VM service (I only have one VM running) "fixes" the problem, so maybe it's not waiting long enough for the VM to stop and it still has files open, leading to the "unclean" tag when the server restarts?

nas-diagnostics-20260603-1648.zipsyslog-192.168.1.5.log

Edited June 3Jun 3 by Free Man

The shutdown timer was exceeded. Time how long it takes to top the array, then add 30 seconds to that and set it in Settings - Disk Settings - Shutdown Timeout.

You might also want to check the timers for shutting down docker containers and VMs are set to values appropriate to your system for those services.

2 hours ago, JorgeB said:

The shutdown timer was exceeded. Time how long it takes to top the array, then add 30 seconds to that and set it in Settings - Disk Settings - Shutdown Timeout.

Wow! Without stopping Docker or VMs first, just hitting "Stop" and "Proceed", it took 3:12 (192 seconds) to stop the array. The miracle is that I've had only 2 unclean shutdowns. I'll move the timeout from 90 to 230 seconds.

48 minutes ago, itimpi said:

You might also want to check the timers for shutting down docker containers and VMs are set to values appropriate to your system for those services.

I did a "Stop All" on the container page and that took 3:06 (186 seconds). I'll definitely add 30 seconds to that timeout as well.

Follow up question, though: The setting (Settings | Docker| Docker Stop Timeout) appears to be the timeout for a single container stop. It's a default of 10, which seems very short to me. Do I want to test shutting down each individual container and set it to a comfort margin above the slowest recorded shut down time, or does this also apply to a bulk shut down, such as on a restart, therefore it needs to be larger than the combined time?

For now, I've bumped the timeout from 10 seconds to 45, assuming my first theory is correct and it's not a total timeout.