rdqmadm (administer replicated data queue manager cluster)

Administer the cluster in a high availability RDQM configuration.


Purpose

Use the rdqmadm command to administer the Pacemaker cluster used in RDQM high availability configurations. (This command is not require for disaster recovery RDQM configurations.)


Syntax

rdqmadm -c-u-s-n nodename-r-n nodename-p-mqmname-p-mqmname-nnodename-p-mqmname-nnodename,nodename-p-mqmname-d


Optional parameters

    -c
    Initialize the Pacemaker cluster, using the settings specified in the /var/mqm/rdqm.ini file. The same command must be run on each of the three nodes by the root user. (We can also run this command as a user in the mqm group if we have configured sudo, see Requirements for RDQM HA solution.) The command fails if the node is already part of a Pacemaker cluster. A node cannot be a member of two Pacemaker clusters.
    -u
    Delete the Pacemaker cluster configuration. The same command must be run on each of the three nodes by the root user. (We can also run this command as a user in the mqm group if we have configured sudo, see Requirements for RDQM HA solution.) The Pacemaker cluster configuration cannot be deleted if any replicated data queue managers (RDQMs) exist.
    -s [-n nodename]
    Suspend the local node (or the specified node if the -n nodename argument is supplied). The command can be run on any of the three nodes by a user in the haclient group, or by root. The node is taken offline. Any replicated data queue managers (RDQMs) running on that node are stopped and restarted on an active node. Queue manager data does not replicate to the offline node. The command fails if the specified node is the last active node.
    -r [-n nodename]
    Resume the local or specified node. The command can be run on any of the three nodes by a user in the haclient group, or by root. The node is brought online. If the node is the preferred location for any replicated data queue managers (RDQMs), the queue managers are stopped and restarted on this node.
    -p -m qmname [-n nodename[,nodename]
    Assign the local or specified node as the Preferred Location for the named queue manager. If the Pacemaker cluster is in a normal state and the Preferred Location is not the current primary node, the queue manager is stopped and restarted on the new Preferred Location. We can specify a comma-separated list of two node names to assign a second preference of Preferred Location.
    -p -m qmname -d
    Clear the Preferred Location so that the queue manager does not automatically return to a node when it is restored.