Auto Maintenance - Additional Information

Maintenance Parameter Store Objects

The Lightbits STS in AWS stores the maintenance state of the instances that are in the progress of maintenance activities in the AWS Parameter Store. The tables below list the parameters and their descriptions.

ReplaceInfo

Field	Description
instanceId	The Instance ID of the instance to be replaced.
forceReplace	The InstanceID of the instance to be forcefully replaced (triggers ‘upgrade’ role), for manually forcing the flow to start.
targetUUID	The UUID of the new server to be configured.
targetInstanceId	The Instance ID of the new server to be configured.
configureRetries	The amount of server configuration retries if the configuration fails.

HealingInfo

Field	Description
instanceId	The Instance ID of the instance to be healed.
targetUUID	The UUID of the new server to be configured.
targetInstanceId	The Instance ID of the new server to be configured.

Scale-Out Information

Field	Description
desiredClusterSize	The N the cluster is to be scaled to.
newServerUUID	The UUID of the new server to be configured.
newServerInstanceID	The Instance ID of the new server to be configured.

Maintenance State Machine States by Role

The tables below list the various maintenance stages an instance can be in.

Healing States

Field	Description
Idle	Set scale in protection for the entire cluster.
scale-out	Scaling out the cluster n+1.
waiting-for-target	Waiting for the target instance to become available for configuration (no action).
waiting-for-ssm	Waiting for the target instance to be available for SSM commands (no action).
configure-target	Configure the target with Ansible.
create-server	Create a target server (replace the node flow).
waiting-for-target-activation	Waiting for the target to become active (no action).
complete-target-activation	Target Node Active, complete lifecycle action to add an instance to NLB.
replace-node	Trigger node replication (source to target).
wait-replace-node	Waiting for node replication to complete (moving volumes to a new node).
disable-server	Disable the source server.
wait-lb-server-disabled	Waiting for the server to be disabled.
delete-server	Delete the source server.
decrease-autoscaling-group	Decrease ASG n-1 (terminates the source instance).
wait-for-asg-decrease	Wait for the source instance to be terminated.
terminate-source	Terminate Source Instance (terminating:wait would end in termination, but this is more optimized).
wait-for-terminate-state	Waiting for the instance to be fully terminated.
complete-replace	Resets scale in protection of instances if replace did not fail.
healing-failed	Replace failed (no action).

Last updated on

Was this page helpful?