On Mon, 2026-01-05 at 04:50 +0000, Andre Robatino wrote: > Without opening the machine, I ran a single pass of memtest86+, which > takes about 80 minutes, no errors. I'm reluctant to do surgery on > this desktop right now as long as it's my primary. When I get a new > desktop and this one becomes secondary, it'll be a lot easier since > it'll be off almost all the time. I'm fairly sure I opened it up > sometime in the last year and would have blown out any dust which > there was not a lot of. > > I strongly suspect some software issue since I only see the > corruption either in the top panel or in the two specific > applications MyPasswordSafe (a very old password manager, I'll have > to migrate eventually) and hexchat, which I stopped running since > there's very little activity on IRC nowadays. Since those two > applications are very old, they might be misbehaving somehow. When I > brought up MyPasswordSafe and saw the dmesg error above, the > MyPasswordSafe window had been corrupted immediately when it started, > so I suspect those are connected. Usually it's either not corrupted > or it happens gradually. I suspect bringing up MyPasswordSafe > triggered the GPU hang.
Could be a specific part of memory, though I'd have thought memtest would have shown up that kind of problem. But I don't think it'd be any good at detecting problems in graphics card memory if you have a separate graphics card with its own RAM. There other programs that are intended for graphics memory testing (research: gpu memtest). If you have *any* plug-in cards on the motherboard, remove and re- insert them. *That* kind of problem (walking cards) occurs by itself over time, through thermal expansion and contraction, and from people picking up PC cases that twist (even slightly) out of shape when moved. Graphics cards with cooling fans can have fan failures, too. I was given a card where the fan had jammed, then melted. I used it for many years with an offboard fan (even bigger than the original) keeping it more than adequately cool. A full cold boot sometimes helps, shut down, pull the power plug out, wait as long as your patience allows, plug in and start up again. I had a system with faulty RAM that *apparently* worked fine for many years, until a new install must have allocated the RAM in a different manner and the fault occurred in some (now) important area that caused crashes. I didn't know it was faulty until the new install crashed often and I ran memtest. One bad bit in somewhere important can easily be fatal, one bad bit somewhere unimportant might not have any noticeable effect. -- uname -rsvp Linux 3.10.0-1160.119.1.el7.x86_64 #1 SMP Tue Jun 4 14:43:51 UTC 2024 x86_64 (yes, this is the output from uname for this PC when I posted) Boilerplate: All unexpected mail to my mailbox is automatically deleted. I will only get to see the messages that are posted to the mailing list. -- _______________________________________________ users mailing list -- [email protected] To unsubscribe send an email to [email protected] Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/[email protected] Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
