Hi Krystian, I've set up the full bisection matrix you requested, with one host per data point:
| Host | Kernel | Commit 1 (c861cac950fb, stop_machine) | Commit 2 (009836b4fa52, migrate_swap) | |---|---|---|---| | srv08 | mainline v6.13.0 | absent | absent | | srv06 | mainline v6.14.0 | present | absent | | srv07 | mainline v6.15.6 | present | absent | | srv05 | mainline v6.15.7 | present | present | All four hosts are up and running with their full VM load (7-16 guest VMs each depending on host). I'll monitor them and report back with uptime/panic results as they come in. For context, srv04 is separately still running mainline v6.8.12 (neither commit) with Viktor's suggested boot parameters (`processor.max_cstate=1 idle=nomwait`) - currently at ~2 days uptime, versus ~3 days before it panicked without those parameters on the same kernel. Will keep watching that one too. I'll let all five run and update you once we have enough data (or a panic) to draw conclusions. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/2163642 Title: Kernel panic "Attempted to kill the idle task" on AMD Opteron multi- node NUMA under KVM (6.8.0-137) To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2163642/+subscriptions -- ubuntu-bugs mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
