node 1 randomly going down

[tejasthakur123]

I have 4 node cluster built on HP consumer-grade computers ( added additional HDD via USB 3.0 interface for CPEH)
one node randomly going down not able to fix this can someone please help ?
Attached the log from node

 

[floh8]

---

Maybe RAM defect

 

[tejasthakur123]

How can I fix this ?

 

[floh8]

First check if RAM is defect

 

[alexskysilk]

kernel version
cpu model/motherboard
kernel boot line

edit: if the other nodes are identical, this one is defective. probably ram but not necessarily.

 

[tejasthakur123]

pveversion
root@pve02:~# pveversion
pve-manager/8.2.2/9355359cd7afbae4 (running kernel: 6.8.4-2-pve)
root@pve02:~#

cpu model/motherboard
> hp elitedesk 800 mini G3

 

[tejasthakur123]

root@pve02:/var/log# cd
root@pve02:~# dmesg | grep pvestatd
[55603.689190] pvestatd[1472]: segfault at 64e245fdd6ec ip 000064e245fdd6ec sp 00007ffe17615fb8 error 15 in perl[64e245fcc000+1b1000] likely on CPU 1 (core 1, socket 0)
root@pve02:~#

 

[justinclift]

Yeah, that sounds like a hardware issue. Probably need to try memtest86+ on that specific box to help figure things out for sure:

https://github.com/memtest86plus/memtest86plus

 

[tejasthakur123]

thanks justinclift