Server Fails to Dump a Kernel Core after NMI Received

This document (000020754) is provided subject to the disclaimer at the end of this document.

Environment

SUSE Linux Enterprise Server 15 All Releases
SUSE Linux Enterprise Server 12 All Releases

Situation

Trying to capture a kernel core dump using the NMI switch, but it does not crash. The following messages are observed in /var/log/messages and dmesg output:

kern.emerg kernel: [71431.090192] Uhhuh. NMI received for unknown reason 20 on CPU 0.
kern.emerg kernel: [71431.090193] Do you have a strange power saving mode enabled?
kern.emerg kernel: [71431.090194] Dazed and confused, but trying to continue

The NMI kernel parameters were configured as:

# sysctl -a | grep -i nmi
kernel.nmi_watchdog = 0
kernel.panic_on_io_nmi = 0
kernel.panic_on_unrecovered_nmi = 0
kernel.unknown_nmi_panic = 0

Resolution

Configure the kernel parameters to allow core dump on NMI.

1. Edit /etc/sysctl.conf and add the following parameters:
# vi /etc/sysctl.conf
kernel.panic_on_io_nmi = 1
kernel.panic_on_unrecovered_nmi = 1
kernel.unknown_nmi_panic = 1

2. Reboot the server or run sysctl -p to apply the changes.

Cause

The kernel parameters did not allow kernel core dump from NMI.

Additional Information

The NMI received for unknown reason is linked directly to the kernel.unknown_nmi_panic parameter. However, setting the other parameters will allow all NMIs to trigger core dumps. If you only want the unknown NMIs to trigger a core dump, then only set kernel.unknown_nmi_panic = 1 in /etc/sysctl.conf.

Disclaimer

This Support Knowledgebase provides a valuable tool for SUSE customers and parties interested in our products and solutions to acquire information, ideas and learn from one another. Materials are provided for informational, personal or non-commercial use within your organization and are presented "AS IS" WITHOUT WARRANTY OF ANY KIND.