UNRAID server randomly reboots while rebuilding

[unethical-hone3689]

Hi,

 

Server has been working for around 2 months without issue. Removed a disk and am now shrinking the array. Upon rebuild the server reboots at random intervals. Sometimes I get to 7%, 15%, 22", 45%, 70% and then I find the machine has rebooted.

 

I have thought it to be a cooling issue and have server panels removed, included a floor fan, and opened windows. Ambient temp is 22C and the hottest drive is 40C during rebuild.

 

it could be a hardware issue but everything is less than 3 months old. PSU is 850W Corsair, Intel i3, 16gb ram, all branded. The LSI HBA card might be over heating too, but hard to say but I've given as much ventilation as possible. 

 

I have one bad disk in cache pool but I don't think that could cause a reboot.

 

 

htpc-vault-diagnostics-20230918-1357.zip

[JorgeB]

If it really reboots on its won, instead crashing or hanging, it's most likely a hardware issue, you can enable the syslog server and post that after a crash, but if it's hardware most likely there won't be anything relevant logged.

[unethical-hone3689]

Short update.

 

Reboot in safe mode and then try to start the array, it performs a read check. All drives connected to HBA are reading errors in the millions. Drives attached to mobo are fine. 

 

Removed HBA (LSI 3100 I think - 16 port) and changed the thermal grease on the chips. They were all completely dry. Reinserted and now performing another read check - no errors this time but parity drives are now disabled. One is Toshiba and the other Seagate Ironwolf, both new.

 

Tried to get syslog from last reboot but the syslog was 4gb is size, tried to d/l but system rebooted again and came with a blank syslog. Changing the settings to have syslog rotation so it should keep more than one now.

 

 

[itimpi]

10 minutes ago, unethical-hone3689 said:

Removed HBA (LSI 3100 I think - 16 port) and changed the thermal grease on the chips. They were all completely dry. Reinserted and now performing another read check - no errors this time but parity drives are now disabled. One is Toshiba and the other Seagate Ironwolf, both new.

If read check is error free then you could simply rebuild parity onto the same drives using the process documented here in the online documentation accessible via the ‘Manual’ link at the bottom of the GUI or the DOCS link at the top of each forum page.

 

Another possibility would be to use Tools->New Config with the Preserve All option.    When you return to the Main tab you could then tick the Parity is Valid checkbox to start the array without immediately rebuilding parity.     However since you have been having problems it is quite possible parity is not completely valid so you would need to then run a correcting parity check to ensure it is valid for the current data drives.

[unethical-hone3689]

That's correct, Parity was never achieved as the server reboots. I'll follow your suggestion and provide an update later.

[unethical-hone3689]

I'm rebuilding parity now.

 

HBA went from 70mb/s to 110mb/s just with new grease. Thanks for the help, will update more later.

[unethical-hone3689]

array seems to be fine now, what a relief.

 

Also, I have added a ssd to the pool (I think this will be automatically placed in RAID 1) but it says the device is disabled. What is the proper way to troubleshoot?

 

Any help here is again appreciated.