Graceful Shutdown of an Openshift 4 Cluster
This SOP should be followed in the following scenarios:
- Shutting down an Openshift 4 cluster.
Steps
Prequisite steps:
- Follow the SOP for cordoning and draining the nodes.
- Follow the SOP for creating an
etcd
backup.
- Get the nodes
nodes=$(oc get nodes -o name | sed -E "s/node\///")
- Shutdown the nodes from the administration box associated with the cluster eg prod/staging.
for node in ${nodes[@]}; do ssh -i <ssh_key> core@$node sudo shutdown -h now; done