My setup is a mini-pc, an Orico 5-bay usb enclosure for the array (4 bays populated) and an Orico 4-bay ssd enclosure for the cache (all bays populated). The system performs perfectly with one exception, I cannot shutdown the array (or reboot, or just turn it off). It gets hung (forever). What I've determined is when I need to do one of those 3 things, I have to "reboot -f". I'm always careful to stop docker/vms before doing this so there is no file activity going on. However, Unraid always detects an unclean shutdown so it always starts a parity check. It's no big deal the turn that off. However, I'm wondering if there's a better way...If the answer is "don't use usb drives" that's not really an option for me, so I might just have to put up with the inconvenience. In any event, is there a config parameter or something else that might fix this?

mini12-diagnostics-20260116-1615.zip

9 hours ago, jrmcmillan said:

Orico 5-bay usb enclosure for the array (4 bays populated) and an Orico 4-bay ssd enclosure for the cache

I think you should narrow down which one cause problem first ( or both ? )

Edited January 17Jan 17 by Vr2Io

I know for sure the main array causes a problem with the shutdown/reboot - e.g., the 5-bay causes a problem because at one point, I had the cache (4-bay ssd) off. I do not know whether the 4-bay is a problem or not.

From my way of thinking, Unraid is looking for something from the disks when shutting down the array which it's not getting, hence it just gets stuck waiting (forever). The only way to recover at that point is to power off the server. As I understand it, "reboot -f" tells Unraid not to look for anything, just reboot - which is why that works - but with the unclean shutdown flag set.

You will always get unclean shutdown if the array can't be stopped before reboot or powerdown. Unraid writes the array started/stopped status to the flash drive when the array is started/stopped. If you boot and it says array was not stopped that is the definition of unclean shutdown.

Another possible "unclean" scenario is if the flash drive can't be written for some reason to update that status.

I get that...the issue is what does Unraid expect that it's not getting from the array (inside the usb enclosure) when trying to shut it down? And, is that something I can configure in Unraid so it doesn't get stuck looking for it?

There is this setting:

You should manually check to see that you can shutdown the array. If you can't, you may have a Docker container or VM that will not release a disk.

It never hits this timer. It gets stuck waiting for something...and I think it has to do with closing files as "shutdown -f" works.

I think you may misunderstand the purpose of that setting. If the array is not stopped by that point, UnraID WILL FORCE a shutdown and that will result in an unclean shutdown.

You should time the time required to stop the array from the button on the MAIN tab to determine how long it takes to stop your array. If it never stops, you can never do a clean shutdown and you have to find what it preventing stopping the array.

20 hours ago, jrmcmillan said:

I'm always careful to stop docker/vms before doing this

Your system share has files on the array. You should clean that up. And unless you disable Docker and VM Manager in Settings, that share might still have open files. Unraid should stop those while stopping the array though.

No, that's not what happens. The array will never stop. It gets hung up and will sit there for a zillion hours, stuck. SSD is inaccessible as well.

22 minutes ago, jrmcmillan said:

No, that's not what happens. The array will never stop. It gets hung up and will sit there for a zillion hours, stuck. SSD is inaccessible as well.

Now for what to you may be the big reveal. If the array does not stop, you can never have a clean shutdown. When that 'Shudown down timer' expires, Unraid will then force-ably shut the system down resulting in an unclean shutdown and require a parity rebuilt on startup.

Circular discussion. I don't want to sound impatient, but you just described what I originally posted. I'm looking for a potential solution, not a restatement of what's already been documented.

You have to find the process that is causing the problem. You can disable the Docker and see if you can stop the array then. You can manually stop any VM's and check again. If neither of those tests don't allow stopping, then you can try booting in the 'SAFE' mode which disables all plugins. Finding the problem process is a tedious procedure of elimination.

Pls further test if start array in maintenance mode then stop array, does this success ？

If still hang, does enclosure activity LED periodic flash ？

Edited January 17Jan 17 by Vr2Io

On 1/16/2026 at 4:27 PM, jrmcmillan said:

If the answer is "don't use usb drives"

Maybe try another enclosure?

On 1/16/2026 at 4:27 PM, jrmcmillan said:

an Orico 5-bay usb enclosure for the array (4 bays populated) and an Orico 4-bay ssd enclosure for the cache

Do you have links for those?

I turned off docker (including table readmore.js) and vms. And, I could stop the array and restart it! So, I guess my procedure going forward is to turn those things off before stopping the array. Thanks for the hint!