MQSeries.net :: View topic

lucac

I have the same problem described in http://www.mqseries.net/phpBB2/viewtopic.php?t=3068&highlight=amqzlaa0+process (amqzlaa0 processes continually increase until the QMGR dies and it has to be restarted), but my Queue Manager is working in an cluster enivronment.

The situation is the follow.

The Queue Manager with the problem is installed on a Redhat Advanced Server 2.1. (that we'll call QM1) and it is a partial repository of the cluster. It shares a queue with an other Queue Manager (that we'll call QM2 and it is a full repository in the cluster) installed on a Redhat 7.2.

Both the Queue managers receive messages into the shared queue from another full repository (installed on OS390), then the same application (installed on both servers) get the messages from the queue.

I cannot understand why on QM1 I had more than 900 amqzlaa0 processes/threads while on QM2 I have just 60 processes/threads. I think I should find approximately the same number of amqzlaa0.

In AMQERR01.LOG on QM1 I find several messages "AMQ9209: Connection to host 'serverofQM2 (10.187.70.104)' closed"; of cours all channels are running properly and I have not network problem.
Could it be the cause of the problem?

Both the MQ installed on Linux (7.2 and Advanced Server) are 5.2 without ptf.

Thanks for any idea or suggestion that can help me to solve the situation.

Thanks.
Luca