Good Day,

 

I am getting random reboots on my server. Most recently at 3.57am this morning when my logs start. Happened twice yesterday also. 

 

Suspect the power supply may be on its way out though noticed an MCE error I cannot interpret.

 

Full diagnostic attached, and extract below. Can anyone more learned than I let me know what is happening here? CPU going?

 

Jan  3 08:03:32 Unraid kernel: mce: [Hardware Error]: Machine check events logged
Jan  3 08:03:32 Unraid kernel: [Hardware Error]: Corrected error, no action required.
Jan  3 08:03:32 Unraid kernel: [Hardware Error]: CPU:0 (17:71:0) MC8_STATUS[Over|CE|MiscV|AddrV|-|-|SyndV|CECC|-|-|-]: 0xdc2040000000010b
Jan  3 08:03:32 Unraid kernel: [Hardware Error]: Error Addr: 0x0000000000003240
Jan  3 08:03:32 Unraid kernel: [Hardware Error]: IPID: 0x000700b020350100, Syndrome: 0x000000432a1f0803
Jan  3 08:03:32 Unraid kernel: [Hardware Error]: L3 Cache Ext. Error Code: 0, Shadow Tag Macro ECC Error.
Jan  3 08:03:32 Unraid kernel: [Hardware Error]: cache level: L3/GEN, tx: GEN, mem-tx: GEN
Jan  3 08:09:00 Unraid kernel: mce: [Hardware Error]: Machine check events logged
Jan  3 08:09:00 Unraid kernel: [Hardware Error]: Corrected error, no action required.
Jan  3 08:09:00 Unraid kernel: [Hardware Error]: CPU:0 (17:71:0) MC8_STATUS[Over|CE|MiscV|AddrV|-|-|SyndV|CECC|-|-|-]: 0xdc2040000000010b
Jan  3 08:09:00 Unraid kernel: [Hardware Error]: Error Addr: 0x0000000000003840
Jan  3 08:09:00 Unraid kernel: [Hardware Error]: IPID: 0x000700b020350100, Syndrome: 0x000000432a1f0803
Jan  3 08:09:00 Unraid kernel: [Hardware Error]: L3 Cache Ext. Error Code: 0, Shadow Tag Macro ECC Error.
Jan  3 08:09:00 Unraid kernel: [Hardware Error]: cache level: L3/GEN, tx: GEN, mem-tx: GEN

 

Many Thanks.

unraid-diagnostics-20240103-0803.zip

Start by taking a look here:

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=819173

 

Thanks for getting back to me @JorgeB.

 

I have changed the BIOS setting as suggested and had another involuntary reboot shortly after restarting.

 

It would be unusual for this to be the problem as the server has been running fine for a few years now and this is only a recent problem.

 

If it provides any further insight, when the server reboots docker is disabled. It ran fine overnight without docker enabled, though when I re-enabled this morning and started using my VM as well it rebooted on its own.

 

Perhaps this is indicative of CPU load rising causing the reboot?

I would think the most likely things to cause a reboot would be thermal (e.g. CPU overheating), or PSU issues.

Thanks for your responses.

 

I rebuilt the server including new thermal paste, a good clean out and improvement of airflow in the case.

 

Haven't had any un scheduled reboots in a week and the MCE error is no longer appearing. Safe to assume the CPU was overheating causing the error & reboots but I did take out older older GTX970 graphics card also - perhaps that had some influence. 

 

Lesson for others reading - clean out your fans/case/heatsinks and ensure good airflow. Not like I did (hadnt been cleaned for 3 or so years).