kdump fails to create core and hangs with "I'm in purgatory"

This document (7022653) is provided subject to the disclaimer at the end of this document.

Environment

SUSE Linux Enterprise Server 11 Service Pack 4 (SLES 11 SP4)

Situation

This is only seen on PowerPC.

After a kernel panic/crash the system hangs and no vmcore is generated.

Last message in the Call Trace on the screen is:

Call Trace:
[c00000076f13faf0] [c000000000499110] .__handle_sysrq+0x170/0x220 (unreliable)
[c00000076f13fbb0] [c000000000499234] .write_sysrq_trigger+0x74/0x80
[c00000076f13fc30] [c00000000026ccc4] .proc_reg_write+0xb4/0x128
[c00000076f13fce0] [c0000000001efa08] .vfs_write+0xf0/0x1d8
[c00000076f13fd80] [c0000000001efc14] .SyS_write+0x5c/0xc8
[c00000076f13fe30] [c00000000000996c] syscall_exit+0x0/0x40
Instruction dump:
54000f7c 7d290378 99280003 ebc1fff0 4e800020 fbc1fff0 ebc2ce80 38000001 
e93e8018 90090000 7c0004ac 39200000 
<98090000> ebc1fff0 4e800020 786b0020 
Sending IPI to other cpus...
I'm in purgatory

Resolution

The kernel update version 3.0.101-108.35.1 released February/27 2018 includes the patch to fix up the kdump base cap to 128M. Please install the recommended update.

Until the kernel can be updated a workaround (offset of 128M) for the crashkernel kernel command line can be used:

crashkernel=size[@offset]
or
crashkernel=<range1>:<size1>[,<range2>:<size2>,...][@offset]

Cause

On PowerPC (PowerNV) the first memory block can be very large, so kdump base is set to 512M, but memory is needed from an area capped at 256M and this results in kdump to fail.

Additional Information

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.