On Sat, Nov 25, 2023 at 2:08 AM Tim <ignored_mailbox@xxxxxxxxxxxx> wrote: > > On Fri, 2023-11-24 at 21:02 -0500, Jeffrey Walton wrote: > > You are having far too many problems with too many programs. The > > common fixes are not helping. > > Remembering tales from long ago - how updating an OS often seemed to > induce faults in hardware that was apparently working fine before. > There's many reasons why this may be the case. > > The newer code is poking different areas of memory than before. > Previously a memory fault existed, but not in an area that you noticed > (maybe one pixel in the display never worked) and the system didn't > throw a tantrum about, but now that bit of RAM is being used by some > important operating system process. I've experienced this kind of > thing (one install worked fine, apparently, even with RAM that tested > as faulty, the next install did not). And, of course, there's things > like manufacturers releasing faulty hardware that fails on a newer > driver thanks to happenstance that the prior driver didn't provoke. > > The install process was CPU intensive for a prolonged time and stressed > something on a system that has spent most of its time the last few > years just idling only. A heatsink wasn't good enough, or clogged up > with fluff, or not fastened properly, or the thermal paste dried out or > wasn't applied well in the first place, or the cooling fan was failing, > and the CPU overheated. Some power filtering component on the edge of > dying, did. A vibrating DVD drive finally weakened some bad soldering > joint... > > And another one I've experienced: Prior to an upgrade, you've picked > up the box and cleaned it, or just moved it. The box isn't rigid, it > twists a bit, and some of the daughter boards aren't properly seated to > the motherboard any more (video cards, memory cards, etc). This also > happens when the ambient temperature changes a lot - things creep out. > Even a CPU socket can go bad, and I'm surprised this doesn't happen > more often, considering the huge cooling devices that are crushed onto > them, these days. As far back as the old Apple ][ it was a common > debugging procedure to re-seat all the cards and socketed chips, to fix > odd failures. Also, some cards didn't have very good edge connectors, > cleaning them often made things better. > > I remember a range of Macs had problems (as they aged) with the factory > soldering not being up to par. People would remove all the plastic > parts and back the motherboard in an effort to re-flow the solder. I > was given one in that condition, but decided it wasn't worth the pain > to try and fix. Solder faults on complex machine boards tend to be > everywhere, not just one spot. Often caused by the solder being too > contaminated on the build day, or wrong temperature, or the boards > being contaminated. Things sometimes just age and fail from cumulative > decay. > > So, come OS install and update times, I tend to open a box, inspect > heatsinks, clean it, and reseat all the connections. Having a spare > power supply to swap over is handy, too. They don't always age well, > especially the bargain basement types. I'm guessing he's got a chip that is overheating when powered on or under load. Maybe due to a bad fan, maybe a bad capacitor, maybe something else. It would be interesting to see an IR image of the motherboard and chips while powered on. See if anything is hitting 70 or 80°C. Jeff -- _______________________________________________ users mailing list -- users@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to users-leave@xxxxxxxxxxxxxxxxxxxxxxx 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/users@xxxxxxxxxxxxxxxxxxxxxxx Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
