Hello unRAID Community

Since a week now my unRAID server has strange behaviour.

In random timeframes it seems to reboot itself and starts a parity check. When it kind of "freezes" all access ist lost, I cannot reach the web gui nor any running docker container.

The system is just not responding and booting again. But not the full machine with a short power cut or full reboot, it seems to just restart unRAID. And then it starts the parity check again.

No matter if I move the date of the check or anything - it always starts. The server will reboot at random occasions no matter if the parity check is ongoing or canceled, if docekrs are running or not or if plex is streaming a video. Completely random.

The Syslog says:

Sep 30 15:41:13 Jukebox kernel: Warning: PCIe ACS overrides enabled; This may allow non-IOMMU protected peer-to-peer DMA
Sep 30 15:41:13 Jukebox kernel: ACPI: Early table checksum verification disabled
Sep 30 15:41:13 Jukebox kernel: floppy0: no floppy controllers found
Sep 30 15:41:13 Jukebox kernel: amdgpu 0000:07:00.0: amdgpu: psp gfx command LOAD_TA(0x1) failed and response status is (0x7)
Sep 30 15:41:13 Jukebox kernel: amdgpu 0000:07:00.0: amdgpu: psp gfx command INVOKE_CMD(0x3) failed and response status is (0x4)
Sep 30 15:41:13 Jukebox kernel: amdgpu 0000:07:00.0: amdgpu: SECUREDISPLAY: query securedisplay TA failed. ret 0x0
Sep 30 15:41:37 Jukebox rc.local: error log : /var/log/graphql-api.log
Sep 30 15:42:08 Jukebox root: error log : /var/log/graphql-api.log
Sep 30 15:42:18 Jukebox root: Error response from daemon: network with name br0 already exists
Sep 30 15:42:18 Jukebox root: Error response from daemon: endpoint with name watchtower already exists in network br0
Sep 30 18:41:09 Jukebox kernel: mce: [Hardware Error]: Machine check events logged
Sep 30 18:41:09 Jukebox kernel: [Hardware Error]: Corrected error, no action required.
Sep 30 18:41:09 Jukebox kernel: [Hardware Error]: CPU:0 (17:60:1) MC17_STATUS[Over|CE|MiscV|AddrV|-|-|SyndV|CECC|-|-|-]: 0xdc2040000000011b
Sep 30 18:41:09 Jukebox kernel: [Hardware Error]: Error Addr: 0x000000068011c040
Sep 30 18:41:09 Jukebox kernel: [Hardware Error]: IPID: 0x0000009600050f00, Syndrome: 0x000004960a400902
Sep 30 18:41:09 Jukebox kernel: [Hardware Error]: Unified Memory Controller Ext. Error Code: 0
Sep 30 18:41:09 Jukebox kernel: [Hardware Error]: cache level: L3/GEN, tx: GEN, mem-tx: RD
Sep 30 19:43:45 Jukebox root: Fix Common Problems: Warning: Docker Application immich has an update available for it
Sep 30 19:43:45 Jukebox root: Fix Common Problems: Warning: Docker Application n8n has an update available for it
Sep 30 19:43:45 Jukebox root: Fix Common Problems: Warning: Docker Application plex has an update available for it
Sep 30 19:43:53 Jukebox root: Fix Common Problems: Error: Machine Check Events detected on your server
Sep 30 19:43:53 Jukebox root: mcelog: ERROR: AMD Processor family 23: mcelog does not support this processor.  Please use the edac_mce_amd module instead.
Sep 30 19:43:56 Jukebox root: Fix Common Problems: Other Warning: dynamix.unraid.net Support Updated ** Ignored
Sep 30 19:43:57 Jukebox root: Fix Common Problems: Other Warning: Unassigned Devices Plus not installed ** Ignored

It seems to display a CPU Error? Is this possible? Which is funny, because the error message is not before the reboot but randomly placed.

I tried switching RAM modules but no change of behaviour.

I attached two diagnostics with a few days inbetween. Can you see more than I do?

I am very new to this kind of troubleshooting and I would appreciate any help.

Thanks in advance

Wooz

jukebox-diagnostics-20250930-1943.zip jukebox-diagnostics-20250929-0105.zip

42 minutes ago, woozaa. said:

reboot itself and starts a parity check

You will get a parity check if it is not shutdown cleanly (array is stopped first).

Is your BIOS configured to boot on power restore?

Hi trurl

Thanks for your reply!

28 minutes ago, trurl said:

You will get a parity check if it is not shutdown cleanly (array is stopped first).

Werent aware of this! Thanks!

27 minutes ago, trurl said:

Is your BIOS configured to boot on power restore?

I will check asap since I cant remember. But most likely it is. I remember to do the setup as secure as possible to take care of any situations when I am not at home.

But I was next to the machine several times when unRAID restarts. There was no power interrupt. It seems as if only the OS itself reboots (is this even possible?).

Edited September 30, 2025Sep 30 by woozaa.

1 hour ago, woozaa. said:

only the OS itself reboots (is this even possible?)

The OS can reboot if it is commanded to reboot. It will not reboot if it crashes.

1 hour ago, woozaa. said:

There was no power interrupt

Possibly a PSU issue then, or maybe a faulty or stuck power button. Are you sure your cat didn't do it? (happened to me!)

Do you have an UPS?