Note
This feature is available starting from the MCP 2019.2.16 maintenance update. Before using the feature, follow the steps described in Apply maintenance updates.
The Kernel Crash Dump (kdump) mechanism allows you to save the system memory
events for later analysis. This section instructs you on how to configure
kernel crash dumping for all or particular nodes. The dump will be saved in the
/var/crash directory. For more details, see Kernel Crash Dump.
To configure kernel crash dumping:
Log in to the Salt Master node.
Open the cluster level of your deployment model.
In <cluster_name>/infra/init.yml, add the following variable:
parameters:
linux:
system:
kernel_crash_dump:
enabled: true
Apply the changes for the required nodes or for all nodes. To apply the changes for all nodes:
salt -C '*' state.sls linux.system.kernel_crash_dump
Manually reboot the nodes for which you have enabled kernel crash dumping. For example, to reboot the OpenStack compute nodes:
salt -C 'cmp1*' system.reboot --async
Once done, you can view the dump in the /var/crash directory.