Hi Krystian,
srv06 (mainline v6.14.0, commit 1 only) panicked after ~39 hours - a
soft lockup warning appeared first ("watchdog: BUG: soft lockup - CPU#24
stuck for 141006s!" on a CPU 2/KVM thread), then escalated to the host
becoming fully unresponsive over SSH.
Notably, srv07 (also mainline v6.15.6, commit 1 only, no commit 2) has
been stable for 9+ days with the identical commit configuration, so this
crash on srv06 doesn't cleanly confirm or rule out commit 2 as necessary
- it's consistent with the high variability we've seen throughout this
investigation.
I've rebooted srv06 back onto v6.14.0, this time also adding Viktor's
boot parameters (`processor.max_cstate=1 idle=nomwait`) - confirmed
active via /proc/cmdline. This gives us a combined data point (commit 1
present, commit 2 absent, boot workaround applied) to compare against
srv04's ongoing test (neither commit, workaround applied, currently 11+
days stable).
Current status of all hosts:
- srv04: v6.8.12 (neither commit) + boot params - 11d 14h, stable
- srv08: v6.13.0 (neither commit) - 10d 15h, stable
- srv07: v6.15.6 (commit 1 only) - 9d 14h, stable
- srv05: v6.15.7 (both commits) - 9d 14h, stable
- srv06: v6.14.0 (commit 1 only) + boot params - fresh restart after the panic
above
--
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
