Monitoring and statistics on IBM MQ for z/OS

IBM® MQ for z/OS® has a set of facilities for monitoring the queue manager, and gathering statistics.

IBM MQ supplies facilities for monitoring the system and collecting statistics. For further information about these facilities, see the following sections:

Online monitoring

IBM MQ includes the following commands for monitoring the status of IBM MQ objects:

DISPLAY CHSTATUS displays the status of a specified channel.
DISPLAY QSTATUS displays the status of a specified queue.
DISPLAY CONN displays the status of a specified connection.

For more information about these commands, see The MQSC commands.

IBM MQ trace

IBM MQ supplies a trace facility that you can use to gather the following information while the queue manager is running:

Performance statistics

The statistics trace gathers the following information to help you monitor performance and tune your system:

Counts of different MQI requests (message manager statistics)
Counts of different object requests (data manager statistics)
Information about Db2® usage ( Db2 manager statistics)
Information about Coupling Facility usage (Coupling Facility manager statistics)
Information about SMDS usage (shared message data set statistics)
Information about buffer pool usage (buffer manager statistics)
Information about logging (log manager statistics)
Information about storage usage (storage manager statistics)
Information about lock requests (lock manager statistics)

Accounting data

The accounting trace gathers information about the processor time spent processing MQI calls and about the number of MQPUT and MQGET requests made by a particular user.
IBM MQ can also gather information about each task using IBM MQ. This data is gathered as a thread-level accounting record. For each thread, IBM MQ also gathers information about each queue used by that thread.

The data generated by the trace is sent to the System Management Facility (SMF) or the generalized trace facility (GTF).

Events

IBM MQ events provide information about errors, warnings, and other significant occurrences in a queue manager. By incorporating these events into your own system management application, you can monitor the activities across many queue managers, for multiple IBM MQ applications. In particular, you can monitor all the queue managers in your system from a single queue manager.

Events can be reported through a user-written reporting mechanism to an administration application that supports the presentation of the events to an operator. Events also enable applications acting as agents for other administration networks, for example NetView®, to monitor reports and create the appropriate alerts.