Machine Check Events and/or parity read errors

[Kudagra]

I'm not sure if these two items are correlated, but I just woke up to a MCE error in fix common problems, and also a disabled parity drive due to read errors.

 

I was having read errors with multiple drives a couple weeks back, so I did extensive RAM testing, but it ended up being resolved (I think?) by moving my HDD sata power plugs to a different PSU rail. Now this single drive comes comes back with more errors (the system disabled it).

 

May 20 08:07:38 Osgiliath kernel: mce: [Hardware Error]: Machine check events logged
May 20 08:07:38 Osgiliath kernel: [Hardware Error]: Deferred error, no action required.
May 20 08:07:38 Osgiliath kernel: [Hardware Error]: CPU:1 (19:21:0) MC12_STATUS[Over|-|-|AddrV|PCC|SyndV|UECC|Deferred|Poison|Scrub]: 0xc765ffc883007f37
May 20 08:07:38 Osgiliath kernel: [Hardware Error]: Error Addr: 0x0000000000000000
May 20 08:07:38 Osgiliath kernel: [Hardware Error]: IPID: 0x0000000000000000, Syndrome: 0x0000000000000000
May 20 08:07:38 Osgiliath kernel: [Hardware Error]: Bank 12 is reserved.
May 20 08:07:38 Osgiliath kernel: [Hardware Error]: cache level: L3/GEN, tx: DATA

 

 

Any ideas?

 

syslog.txt

Edited May 23, 2023 by Kudagra

[Kudagra]

Any suggestions?

[JorgeB]

The MCE appears to be CPU related, for the disk errors please post the complete diagnostics after some errors.

[Kudagra]

10 hours ago, JorgeB said:

The MCE appears to be CPU related, for the disk errors please post the complete diagnostics after some errors.

Thank you for the response. Would you suggest I test for CPU failure?

 

I've attached a zip of the full diagnostics (which was downloaded at the same time as the previously attached syslog file).

diagnostics-20230522-1711.zip

[JorgeB]

12 hours ago, Kudagra said:

Would you suggest I test for CPU failure?

Stop overclocking RAM, it's a known issue with Ryzen, if it doesn't help try with a new CPU if possible.

 

Disk errors look more like a power/connection problem, replace both cables for that disk.

[Kudagra]

On 6/1/2023 at 12:21 AM, JorgeB said:

Stop overclocking RAM, it's a known issue with Ryzen, if it doesn't help try with a new CPU if possible.

 

Disk errors look more like a power/connection problem, replace both cables for that disk.

 

I disabled the RAM overclocking, but unfortunately I'm still receiving the 'Machine Check Events' error (most recent diag attache). Unfortunately I don't have a second CPU on hand to try, so you have any suggestions on how I can test the health of this CPU?

 

As for the disk errors (for anyone else seeing this problem)- I resolved this with a new PSU. Apparently my previous PSU had one of it's rails failing.

diagnostics-20230808-0941.zip

[JorgeB]

15 minutes ago, Kudagra said:

any suggestions on how I can test the health of this CPU?

Not really, best bet would be using a different one, also look for a BIOS update but that shouldn't do much for this.

[Kudagra]

On 8/11/2023 at 10:09 AM, JorgeB said:

Not really, best bet would be using a different one, also look for a BIOS update but that shouldn't do much for this.

Unfortunately BIOS is already up to date, but I did have this error popup in my log. Think it's related?

 

 

 

Edited February 7 by Kudagra

[JorgeB]

Possibly.

[Kudagra]

Just to provide an update on this problem- it ended up being a faulty CPU as JorgeB suggested.

 

I successfully RMA'd my 5950x a few weeks back (meaning it failed their testing), and I haven't seen the errors since installing the replacement CPU.