get_swap_device: Bad swap file entry

[HPmoss]

I've noticed my Proxmox installation freezing or crashing for past couple of days.

On one occasion I noticed following message at login prompt:

get_swap_device: Bad swap file entry 4003fffffda6dac7

IS it bad RAM module issues or something else? IF it is RAM related them how do I identify faulty module? The bare-metal (HP Z230) has builtin diagnosic, and that cam out all clear when I did RAM test. However, the Memtest+ (within Proxmox when at GRUB) should memory errors while testing.
Machine boots up well and all VM start as normal but after "some time" the bare-metal Proxmox goes into unresponsive mode. Console or SSH is inaccessible. There's no ping reply.

Any clue would be appreciated. Thanks!

 

[HPmoss]

Proxmox version:
pve-manager/8.2.7/3e0176e6bb2ade3b (running kernel: 6.8.12-2-pve)


Some extract from kern.log
kernel: [72785.829595] general protection fault, probably for non-canonical address 0xffff9a13
a58310: 0000 [#1] PREEMPT SMP PTI
2024-10-07T10:08:08.036216+01:00 pve01 kernel: [72785.829665] CPU: 0 PID: 229570 Comm: iou-wrk-1374 Tainted: P O 6.8.
12-2-pve #1

 

[BobhWasatch]

Do you not believe the memtest result? It is a more thorough test than most BIOS provide. Plus, your symptoms sure could be caused by bad RAM. I would get a new RAM module and start swapping until memtest passes.

 

[psilan]

I have the same issue. I suspect it is a memory problem and will test shortly.
Did you ever resolve this?

 

[HPmoss]

It was memory problem at my end BUT it was in fact dust accumulation within the box. After a deep and thorough clean up (using a can of compressed air) I was able to revive the box without any hardware failure.

 

[psilan]

Mine was a very faulty dimm module in the end. Thanks for the reply .