I’m having trouble finding something. I have a rockpro64 running Manjaro. It has KDE on it but for the most part just hosts files, jellyfin, and homebridge. I saw it had an uptime of about 16-18 hours and it seemed to be running out of ram. plasmashell had a nasty memory leak. So I killed the desktop environment and ran it without to see if the problem went away.
It only lasted a few days before I had to power cycle it again. I’m not seeing anything weird in journalctl. What should I be looking for?
and post some more information so we can see what’s really going on. Now we know the symptom of the disease, but we need some more probing to know where the origin lies…
An inxi --admin --verbosity=7 --filter --no-host --width would be the minimum required information… (Personally Identifiable Information like serial numbers and MAC addresses will be filtered out by the above command)
Well, while Plasmashell in general is pretty performant, it does tend to use more RAM as it’s been running.
Your issue, I think, is that you are also running Jellyfin, which is much more of a RAM hog, than Plasma is.
Jellyfin and Plasma can separately run on the RockPro64 for weeks (I know, since I do it) without running out of RAM, but if you run both, I would suspect it runs out quicker.
To me, that does not look like it is a memory error or an out of memory error. It looks more like a either a scsi driver bug or a hardware fault. Still 2G of free and no swap at all used by looking at what is left of the htop screen.
That is defiantly what I would do then. The kernel could very well be the issue not meshing with other program’s libs and at the same time there can be issues with the other programs lib’s so best to do a full system upgrade to try to alleviate any possible underlining issues that may be causing the kernel panic.