This is generally not what a watchdog user. The default Linux kernel behavior upon panic is to print a kernel tombstone and loop forever. It is incumbent upon the user to have properly configured the system for kdump.
This is to allow for a crash dump to be collected. Upon receipt of an NMI from the iLO, the hpwdt driver will initiate a panic. You can check whether your kernel supports it by checking whether there is /proc/sys/kernel.
(Note: The latest kernel introduces a new kernel parameter kernel.hardlockup_panic. The format of the kernel parameter is "=". See full list on kernel.nmi_watchdog: Define whether to enable nmi watchdog and whether hard lockup causes panic. panic(cpu 2 caller 0xffffff800817e634): "Spinlock acquisition timed out: lock=0xffffff8019888a38. mp_kdp_enter(): 2, 3, 4 TIMED-OUT WAITING FOR NMI-ACK, PROCEEDING.It also saves registers for the crash dump on x86.: cache level: L1, tx: DATA, mem-tx: DRD : Machine Check: Invalid Kernel panic - not syncing: Fatal machine check on current CPU Shutting down cpus with NMI Here is some additional logging that I found: Setup the system to use kdump with systems lockups (NMI) and OOM (Out Of Memory) scenarios - Append kernel option “nmi_watchdog=1” in nf – Add following kernel parameters in sysctl: - kernel.unknown_nmi_panic=1 - kernel.panic_on_unrecovered_nmi=1 - vm.panic_on_oom=1 Test by crashing the system intentionally commit 1717f2096b54 ("panic, x86: Fix re-entrance problem due to panic on NMI") and commit 58c5661f2144 ("panic, x86: Allow CPUs to save registers even if looping in NMI context") introduced nmi_panic() which prevents concurrent/recursive execution of panic().
0 kommentar(er)
