IBM MQ Replicated Data Queue Manager (RDQM) High Availability

Essential vocabulary terms and core concepts covering IBM MQ Replicated Data Queue Manager (RDQM) high availability architecture, node roles, quorum, and cluster components.

RDQM (replicated data queue manager)

A high availability solution available on Red Hat Enterprise Linux for x86-64 platforms, consisting of three servers configured in an HA group where one instance synchronously replicates data to the other two instances.

Primary

The server in an RDQM HA configuration that is actively running the queue manager and synchronously replicating its data to the other two servers.

Secondary

The designation for each of the two servers in an RDQM HA group that is not actively running the queue manager but maintains replicated data.

Split-brain

A scenario where broken connectivity causes multiple nodes to run the queue manager independently at the same time, resulting in two conflicting versions of data that require manual intervention to resolve.

Quorum

A state established when nodes can communicate with at least one other node in a three-node system; queue managers can only run on a node that possesses it.

Pacemaker

The cluster manager software that handles the HA group configuration and node monitoring of the three nodes in an RDQM architecture.

DRBD

The component that handles data replication between the three nodes in an RDQM high availability system.

Floating IP address

An optional shared IP address among queue manager instances that allows clients to connect using a single IP address regardless of which node is currently hosting the running queue manager.

Preferred Location

A configuration setting that identifies the specific node where an RDQM should run whenever that node is available.

crtmqm

The control command used to create a high availability replicated data queue manager (RDQM).