SLES server running on VMware ESX is hung and unable to get a kernel core for debugging
This document (7008844) is provided subject to the disclaimer at the end of this document.
SUSE Linux Enterprise Server 11
The nmi_watchdog kernel boot parameter does not work either (the kernel nmi watchdog driver doesn't find the same watchdog timer sources in a VMware ESX guest that it finds on a physical server so the watchdog timer reports an error when initializing in a ESX guest).
Set the SLES server up to gather a kernel core. TID 3374462"Configure kernel core dump capture"
After the server has been rebooted to make the crashkernel active then enter:
sysctl -w kernel.unknown_nmi_panic=1
server:~ # sysctl -w kernel.unknown_nmi_panic=1
kernel.unknown_nmi_panic = 1
From the VMware ESX host machine. Do "vm-support -x" that should give you the VMware ID's for each virtual machine.
/sbin # vm-support -x
VMware ESX Support Script 1.33
Available worlds to debug:
wid=13820675 sles11sp1-server <--used in example
Then do "/usr/lib/vmware/bin/vmdumper 13820675 nmi" Replace 13820675 with your VMware ID.
/sbin # /usr/lib/vmware/bin/vmdumper 13820675 nmi
Sending NMI to guest...
Core will now be generated on the virtual machine for debugging, by default the virtual machine will reboot when finished and core is in /var/crash on SLES 11 and /var/log/dump on SLES 10.
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.
- Document ID:7008844
- Creation Date: 17-Jun-2011
- Modified Date:03-Mar-2020
- SUSE Linux Enterprise Server
For questions or concerns with the SUSE Knowledgebase please contact: tidfeedback[at]suse.com