kdump kernel: Kernel panic - not syncing: Attempted to kill init!

This document (7023399) is provided subject to the disclaimer at the end of this document.

Environment

SUSE Linux Enterprise Server 12 Service Pack 3 (SLES 12 SP3)

Situation

The kdump kernel on shared LPAR may crash.

After a kernel crash booting the kdump kernel and trying to save the crash dump, results in the following failure:

Unable to handle kernel paging request for data at address 0x00000060
Faulting instruction address: 0xc0000000086ad748
Oops: Kernel access of bad area, sig: 11 [#1]
SMP NR_CPUS=2048 
NUMA 
[....]
---[ end trace 026c7350cdb92d81 ]---

Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b

---[ end Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b

With "numa=off" the dump capture is successful and the dumpfile is saved to KDUMP_SAVEDIR.

Resolution

Workaround:

Disable NUMA in the kdump kernel:

1. Append "numa=off" to KDUMP_COMMANDLINE_APPEND in /etc/sysconfig/kdump

2. Restart kdump service

- systemctl restart kdump.service

The kdump update, kdump-0.8.16-7.17.1 released October 2018 includes the patch to turn off NUMA in the kdump kernel.

Cause

After recent fixes targeted at performance improvements in different NUMA configurations, the kdump kernel is failing to initialize NUMA topology in a restricted memory environment.

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.