Protect nodes from automatic restart

Shturval provides a mechanism for automatically restarting nodes in cases of changes in node configuration (NCI) or cluster updates. The drain and reboot policy is set by the shturval-node-config operator. When the node is automatically restarted, the following actions are performed:

• Cordon;
• Drain;
• Reboot.

If necessary, you can protect cluster nodes from automatic restart using special labels. This is applicable for cases when some operations must be performed before draining or rebooting the cluster node.

The sequence of actions will be as follows:

1. Set a ban on drain nodes.
2. Make the necessary changes.
3. Check that the node configuration has received the condition reboot required.
4. Set the permission to apply changes without a drain node over the prohibition on drain.
5. Wait until the node reboots or the configuration is applied.
6. Remove permission to apply changes without drain node.

Commands must be executed using the node name instead of my-special-node.

1. To disable the Drain node, set the node.shturval.tech/disable-drain label using the command:

kubectl label node my-special-node node.shturval.tech/disable-drain=

2. To check the installation of the label, run the command:

kubectl get node my-special-node -o jsonpath='{.metadata.labels}' | jq -r

3. Set the permission to apply the changes using the node.shturval.tech/apply-as-is label.

kubectl label node my-special-node node.shturval.tech/apply-as-is=

4. Check that the resource NodeConfig of the node (nodeconfig) is applied and has the status ready=true:

kubectl get nodeconfig my-special-node -o jsonpath='{.status.ready}'

5. Be sure to remove the node.shturval.tech/apply-as-is label from the node to disable permission to unconditionally apply changes to the node by running the command:

kubectl label node my-special-node node.shturval.tech/apply-as-is-