GPU keeps dissapearing

[LeMihu]

Hi all,

I have a proxmox server with a A380 GPU that im passing to LXCs like Plex and Fileflows.
For the past 9-10 months eveything worked perfectly but for about 2 weeks now it keeps dissapearing.
Like everything is working fine and randomly the LXCs that have it go offline and cant start because the /dev/dri is missing.
If i use lspci the GPU is not there.
After reboots sometimes its ok sometimes it starts without seeing it.
What i observed is that the fan is going 100% when im not seeing it.

Any help would be the most welcomed.

 

[dcsapak]

can you check/post the journal/dmesg of the host?

 

[LeMihu]

For journalctl can you give me a command to filter some of the text? There are over 1 milion lines in it and i cant upload it

 

[dcsapak]

you should be able to gzip it , e.g. by piping it into gzip

journalctl  <your journalctl parameters go here> | gzip > journal.log.gz

a good start would be to use the '--since' and '--until' flags to limit the journal to a specific timerange

an example could be

journalctl --since 2026-03-31 | gzip > journal.log.gz