Restoring a queue manager to its pre-cluster state

When a queue manager is removed from a cluster, it retains knowledge of the remaining cluster members. This knowledge eventually expires and is deleted automatically. However, if you prefer to delete it immediately, you can use the steps in this topic.

Before you begin

It is assumed that the queue manager has been removed from the cluster, and is no longer performing any work in the cluster. For example, its queues are no longer receiving messages from the cluster, and no applications are waiting for messages to arrive in these queues.

Important: If you remove a queue manager from a cluster and refresh it using REPOS(YES), you will not be able to add it back in again by simply altering its CLUSRCVR's CLUSTER attribute. After altering its CLUSRCVR's CLUSTER attribute to be nonblank (that is, the clustername), you will additionally need to issue refresh cluster with REPOS(NO), at which point the internal sequence numbers on the CLUSRCVR will be brought up to date. Then the queue manager will be successful in re-introducing itself to the Full Repositories and the rest of the cluster members. (Note the REPOS(NO) version of the command must be run after the CLUSRCVR channel has been given the correct cluster name.)

This restriction applies to IBM® WebSphere® MQ Version 7.5 only.

About this task

When a queue manager is removed from a cluster, it retains knowledge of the remaining cluster members for up to 90 days. This can have system benefits, particularly if the queue manager quickly rejoins the cluster. When this knowledge eventually expires, it is deleted automatically. However, there are reasons why you might prefer to delete this information manually. For example:

You might want to confirm that you have stopped every application on this queue manager that previously used cluster resources. Until the knowledge of the remaining cluster members expires, any such application continues to write to a transmit queue. After the cluster knowledge is deleted, the system generates an error message when such an application tries to use cluster resources.
When you display status information for the queue manager, you might prefer not to see expiring information about remaining cluster members.

This task uses the INVENTORY cluster as an example. The LONDON queue manager has been removed from the INVENTORY cluster as described in Removing a queue manager from a cluster. To delete knowledge of the remaining members of the cluster, issue the following commands on the LONDON queue manager.

Procedure

Remove all memory of the other queue managers in the cluster from this queue manager:
```
REFRESH CLUSTER(INVENTORY) REPOS(YES)
```

Monitor the queue manager until all the cluster resources are gone:

DISPLAY CLUSQMGR(*) CLUSTER(INVENTORY)
DISPLAY QCLUSTER(*) CLUSTER(INVENTORY)
DISPLAY TOPIC(*) CLUSTER(INVENTORY)