Cluster bootstrap process

Bootstrapping refers to getting a cluster node up and running. When a cluster node recovers from failure, checks are made to keep the node state consistent with the rest of the nodes in the cluster. When you encounter an unresponsive primary node or member node, take following actions:

• You troubleshoot or fix the errors using the various methods that are documented in this IBM Knowledge Center.
• Restart the primary node or the member node virtual appliance.
  • Synchronize the member node with the primary node to update the node with any virtual appliance configuration changes in the cluster.
  • This node becomes stand-alone node if this node is removed from the cluster definition.

The following actions are done by the cluster bootstrap process:

• When a primary node recovers from a failure and detects that there are no changes in the cluster, the primary node does not display any notifications.

• When a member node recovers from a failure, and detects that it continues to be part of the original cluster, synchronization might be needed if the virtual appliance configuration changes were made.

• When a primary node or a member node recovers from a failure and cannot connect with any of the previously known cluster members because the virtual appliance password changed or the primary node is down, the current node is made stand-alone node. A reconnect notification is set.

Parent topic: Troubleshooting virtual appliance problems