I've recently gone from a dual E5-2670 build to an i5-13500 on an Asus Pro WS W680-ACE with Kingston ECC DDR5 RAM. There are a total of 16 SATA HDDs, two SATA SSDs and an M.2 with a five year old EVGA 650 G2 PSU (22A 5V rail).

 

The new build is unstable and seems to go "offline" every 1-2 days. I have:

 

1. Tested the memory with Memtest86 for 24 hours, which passed.

2. Upgraded the BIOS.

3. Removed my UPS to exlcude that as the problem.

 

Two days ago, I was sitting next to the server, and it rebooted out of the blue, and wouldn't go past the BIOS splash screen without shutting down.

 

I'm thinking this could be my PSU dying OR the CPU going to sleep (no idea why), but does anyone see anything in the diagnostics attached to suggest otherwise?

tower-diagnostics-20230325-2004.zip

Edited March 25, 20233 yr by topherino

Enable the syslog server and post that after a crash but if it's a hardware problem unlikely that anything relevant will be logged.

2 hours ago, JorgeB said:

Enable the syslog server and post that after a crash but if it's a hardware problem unlikely that anything relevant will be logged.

 

I've had it running for a few days. The last crash was between Mar 25 16:05:43 and Mar 25 20:02:13. You can see the UPS comms warning every ten minutes leading up to that and then nothing, until I do a hard reboot.

syslog-10.0.0.9.log

There's nothing before the last crash, but there are earlier macvlan call traces, macvlan call traces are usually the result of having dockers with a custom IP address and will end up crashing the server, switching to ipvlan should fix it (Settings -> Docker Settings -> Docker custom network type -> ipvlan (advanced view must be enabled, top right)).