On 2/1/2025 at 1:38 AM, Veah said:

Yes, Yes, it does not hurt.

Making your own thread helps keep issues separate and less confusing for anyone trying to help as well as the individuals seeking it.

Also ensure syslog is enabled and post your diagnostics in your new thread.

both of whicg are done on my post above.... Valuable input.

 

It was a question directed @JorgeBwho seenmd to have running knowledge of the problem

 

@itimpi Thanks for the suggestion, looks llike somethine Dynamix should have dev'd back in the day and just always been installled for me. Is now thanks toy ou!

6 hours ago, Nasha said:

It was a question directed @JorgeBwho seenmd to have running knowledge of the problem

Server rebooting by itself is almost always a hardware issue, but take a look here to see if it helps:

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/page/2/#findComment-819173

 

4 minutes ago, JorgeB said:

Server rebooting by itself is almost always a hardware issue, but take a look here to see if it helps:

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/page/2/#findComment-819173

 

I find it rather curious that it coincided with my alteration of the bios (changed bifurcation from 8+4+4 to Auto), and the fact that that post references changing bios settings.... I might try and revert those settings first, see if that changes anything. I'll report back here what if anything, the fix ends up being..

 

As always, thank you Jorge. Your assistance over the years has been invaluable 

Just a quick update here.  Since the last report, it still did not crash or reboot.  The only thing I did was (started docker) but did not start any containers, so possibly one of the containers must have been the issue but I will wait till this weekend and then slowly start one container and see how it behaves

Hi, I’ve been experiencing unclean shutdowns lately. I thought the issue was due to an old drive, which I have now replaced, but the server still reboots randomly every few days.

For example, it had an unclean shutdown yesterday evening (about 20 horus ago). I’ve attached the diagnostic file, if anyone could help me solve this issue, I would really appreciate it!

Thanks

byo02-diagnostics-20250617-0944.zip

The server rebooting itself is nearly always hardware with commonest culprits being bad power or bad RAM

The syslog in the diagnostics is the RAM version that starts afresh every time the system is booted. You should enable the syslog server (probably with the option to Mirror to Flash set) to get a syslog that survives a reboot so we can see what leads up to the problem. The mirror to flash option is the easiest to set up (and if used the file is then automatically included in any diagnostics), but if you are worried about excessive wear on the flash drive you can put your server's address into the remote server field and log to a share instead..

It is quite wild that the "Unclean shutdown detected" doesn't contain the reason concluded (the code obviously comes to some deduction about making the notification)... presumably it just defaults on leaving some unclean-shutdown temp file which it picks up on next boot.

Parity-check for terabytes takes a reeeeally long time...

Because the manual says a lot of things towards debuging https://docs.unraid.net/unraid-os/troubleshooting/unclean-shutdowns/ but none of them are implemented as detection... just full of speculation.

Why allow shutdowns at all (without force) if there are items detectable by "Dynamix Stop Shell".

For example, having terminal sessions open is hardly a reason to not reboot cleanly, all child processes to terminal sessions are indicated and can be killed at will since we are already going down. The jump conclusion that parity is invalid is equally wild, nice failsafe though.

Why suggest a narrow "/Settings/DiskSettings/Shutdown time-out" of mere seconds if something is wrong, perfect opportunity to inspect the issue, let it simmer since you have something broken anyway. If someone wants to force a reboot then they can do that.

And apprently shutdown is allowed during scrubbing (no warning), quite certainly should be a blocker...

"Unmounting disks...Retry unmounting disk share(s)" show the timeout, indicate triggering the timeout, otherwise it just looks like a latent delay and clean shutdown...

12 minutes ago, Samsonight said:

It is quite wild that the "Unclean shutdown detected" doesn't contain the reason concluded (the code obviously comes to some deduction about making the notification)... presumably it just defaults on leaving some unclean-shutdown temp file which it picks up on next boot.

If the server crashed due to a hardware issue then there would typically be no chance to write out a reason. The system detects the unclean shutdown by the fact that the array was started but then never stopped and probably no shutdown even attempted.

6 minutes ago, itimpi said:

If the server crashed due to a hardware issue then there would typically be no chance to write out a reason. The system detects the unclean shutdown by the fact that the array was started but then never stopped and probably no shutdown even attempted.

Please, never use the words fact and probably in the same sentence, just dont. ❤️

"If", this is the wonderful world of Unraid speculation I just mentioned.

To be honest, then mention the speculative hardware issue as the default message.

To actually address this works on my computer dance, place a detection file for hardware issue crashes and wipe it as the absolute last step of a non-hardware issue restart (one that happens even if not a clean restart). Then we have a much clearer state of conclusions.

Edited June 17, 20251 yr by Samsonight
probably a fact

1 hour ago, itimpi said:

The server rebooting itself is nearly always hardware with commonest culprits being bad power or bad RAM

The syslog in the diagnostics is the RAM version that starts afresh every time the system is booted. You should enable the syslog server (probably with the option to Mirror to Flash set) to get a syslog that survives a reboot so we can see what leads up to the problem. The mirror to flash option is the easiest to set up (and if used the file is then automatically included in any diagnostics), but if you are worried about excessive wear on the flash drive you can put your server's address into the remote server field and log to a share instead..

Thanks. I’ve enabled the syslog server and the option to mirror to flash. When the next shutdown occurs, I’ll post the updated diagnostics.

31 minutes ago, Samsonight said:

Please, never use the words fact and probably in the same sentence, just dont. ❤️

I think my 'if' and 'fact' were both valid ☺️ They were in separate sentences at least.

What is not done (and I think this is what you are suggesting) is recording the fact that a shutdown was initiated but not completed and timeouts started kicking in to allow easier detection of this case. It would enable giving a different message on start from the case where the system suddenly crashes out of the blue. Maybe it is something that Limetech might be persuaded to look into.

Oops wrong message I replied to

Edited June 17, 20251 yr by Nasha

2 hours ago, byographic said:

Hi, I’ve been experiencing unclean shutdowns lately. I thought the issue was due to an old drive, which I have now replaced, but the server still reboots randomly every few days.

For example, it had an unclean shutdown yesterday evening (about 20 horus ago). I’ve attached the diagnostic file, if anyone could help me solve this issue, I would really appreciate it!

Thanks

byo02-diagnostics-20250617-0944.zip

I see you're running AMD CPU... Known issues there. See this post for resolution.
CA Fix Common Problems plugin would pickup whether or not it was the cause as well. However the link it provides for the solution is somewhat deprecated/doesn't specifically include the known issue with AMD. I requested a change a while back, not sure if it's been implemented.

4 hours ago, Nasha said:

I see you're running AMD CPU... Known issues there. See this post for resolution.
CA Fix Common Problems plugin would pickup whether or not it was the cause as well. However the link it provides for the solution is somewhat deprecated/doesn't specifically include the known issue with AMD. I requested a change a while back, not sure if it's been implemented.

I couldn't find the post link 😅

11 minutes ago, byographic said:

I couldn't find the post link 😅

C'mon bro, "the post". We don't ask that kind of clarifying questions. Focus on the flaunting, not the sharing. Known issues there. 😎

Edited June 17, 20251 yr by Samsonight

37 minutes ago, byographic said:

I couldn't find the post link 😅

Crap my apologies. I originally replied to the wrong person and in the copy paste frenzy the link didn't make it.

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/page/2/#comment-819173

Eeek! Something not good. Lost a drive "unmountable: wrong or no file system".

Making a new thread for my quesiton \ attempted fix

Thanks

Fixed with a "mount in maintence mode" and then run a disk check on the one sick disk

That was a useful learning phase. Phew

In case anyone else reads this: https://docs.unraid.net/unraid-os/manual/storage-management/#checking-a-file-system

Edited June 20, 20251 yr by Batter Pudding

On 6/17/2025 at 10:29 AM, itimpi said:

The server rebooting itself is nearly always hardware with commonest culprits being bad power or bad RAM

The syslog in the diagnostics is the RAM version that starts afresh every time the system is booted. You should enable the syslog server (probably with the option to Mirror to Flash set) to get a syslog that survives a reboot so we can see what leads up to the problem. The mirror to flash option is the easiest to set up (and if used the file is then automatically included in any diagnostics), but if you are worried about excessive wear on the flash drive you can put your server's address into the remote server field and log to a share instead..

Hello I've activated the syslog server as indicated.
and this morning roughly 3 hours ago the server rebooted again I've attached the new diagnostic.

thanks

byo02-diagnostics-20250623-0943.zip

9 hours ago, byographic said:

Hello I've activated the syslog server as indicated.
and this morning roughly 3 hours ago the server rebooted again I've attached the new diagnostic.

thanks

byo02-diagnostics-20250623-0943.zip

Did you perform the changes in the linked post?

On 6/23/2025 at 7:11 PM, Nasha said:

Did you perform the changes in the linked post?

I did it today, I've updated the bios and changed - "Power Supply Idle Control" (or similar) and set it to "typical current idle" -

I've also installed the NUT plugin to monitor the UPS which was already present but not monitored through software.

lets hope it doesn't happen again

I had a few unclean shutdowns lately, are these errors any issues to be worried about?

other errors in the logs

also attached the full diagnostics zip.

Any help is greatly appreciated. I would like to figure out what the issue is so I can resolve it.

Thanks,
Deerow

batcomputer-diagnostics-20250714-2201.zip

I just discovered this post in 2025, and have FINALLY managed to avoid unclean shutdowns. I was definitely not clear on the fact of the shutdown timers being independent, so thank you very much!

2 hours ago, molasses2 said:

I just discovered this post in 2025, and have FINALLY managed to avoid unclean shutdowns. I was definitely not clear on the fact of the shutdown timers being independent, so thank you very much!

Hello!

would you be so kind to share your methods on you resolving your issues?

btrfs scrub status /mnt/diskX

is still not cancelled during shutdown, nor does it prevent shutdown / restart. (v7.2.0)

btrfs scrub cancel /mnt/diskX

nor

btrfs balance status /mnt/diskX

Edited November 6, 2025Nov 6 by Samsonight
version

Attach Diagnostics ZIP to your NEXT post in this thread