Yakuza965
Members
-
Joined
-
Last visited
Solutions
-
Yakuza965's post in Silent server hangs and drops of network since 7.3.1 was marked as the answerI eventually tracked this down - as you predicted - to a hardware failure. Specifically, two cores on my CPU (an Intel 14500) that were failing to respond and bringing down the whole system. I managed to capture the efi-pstore DMESG error messages, which included lines like this:
<0>mce: CPUs not responding to MCE broadcast (may include false positives): 4-5 <0>mce: CPUs not responding to MCE broadcast (may include false positives): 4-5 <0>Kernel panic - not syncing: Timeout: Not all CPUs entered broadcast exception handler <0>Shutting down cpus with NMI
I manually disabled those specific cores, which made the system stable again for 5 days where it had previously silently frozen every couple of days. Then I replaced the CPU entirely. The system has been stable for 2 weeks and the old CPU is with intel for replacement under warranty.
Sharing in case it helps others who have trouble with Intel CPUs down the line Thanks for the help in tracking it down.