AMQSCLM: Troubleshooting

The following sections contain information about scenarios that might be encountered while using the sample. Information about potential explanations for a scenario, and options on how to resolve it, are provided.

Scenario: AMQSCLM is not starting

Potential explanation: Incorrect syntax.

Action: Check standard error output for correct syntax

Potential explanation: Queue manager is not available.

Action: Check the report file for message ID CLM0010E.

Potential explanation: Cannot open or create report file or files.

Action: Check standard error output for error messages during initialization.

Scenario: AMQSCLM is not changing a queue to ACTIVE or INACTIVE

Potential explanation: The queue is not in the list of queues to be monitored

Action: Check the -q and -f parameter values.

Potential explanation: The queue is not a local queue in the correct cluster.

Action: Check that the queue is local and in the correct cluster.

Potential explanation: AMQSCLM is not running for this queue manager and cluster.

Action: Start AMQSCLM for the relevant queue manager and cluster.

Potential explanation: The queue is left INACTIVE, CLWLPRTY =0, because it has no consumers. Alternatively, it is left ACTIVE CLWLPRTY >=1, because it has at least 1 consumer.

Action: Check whether consuming applications are attached to the queue.

Potential explanation: The command server of the queue manager is not running.

Action: Check the report files for errors.

Scenario: Messages are not being routed around INACTIVE queues

Potential explanation: Messages are put directly to the queue manager that owns the inactive queue, and the CLWLUSEQ value of the queue is not ANY, and the -u argument is not being used for AMQSCLM.

Action: Check the CLWLUSEQ value of the relevant queue manager, or ensure that the -u argument is used for AMQSCLM.

Potential explanation: There are no active queues on any queue managers. Messages are evenly workload balanced across all inactive queues until a queue becomes active.

Action: Check the status of the queues on all queue managers.

Potential explanation: Messages are put to a different queue manager in the cluster to the one that owns the inactive queue, and the updated CLWLPRTY value of 0 is not propagated to the queue manager of the putting application.

Action: Check that the cluster channels between the monitored queue manager, and the full repository queue manager, are running. Check that the channels between the putting queue manager, and the full repository queue manager, are running. Check the error logs of the monitored, putting, and full repository queue managers.

Potential explanation: The remote queue instances are active (CLWLPRTY=1), but messages cannot be routed to those queue instances because the cluster sender channel from the local queue manager is not running.

Action: Check the status of the cluster sender channels from the local queue manager to the remote queue manager, or managers, with an active instance of the queue.

Scenario: AMQSCLM is not transferring messages from an inactive queue

Potential explanation: Message transfer is not enabled ( -t ).

Action: Ensure that message transfer is enabled ( -t ).

Potential explanation: The queue is not in the list of queues to be monitored.

Action: Check the -q and -f parameter values.

Potential explanation: AMQSCLM is not running for this, or other queue managers in the cluster, that own instances of the same queue.

Action: Start AMQSCLM.

Potential explanation: The queue has CLWLUSEQ = LOCAL or CLWLUSEQ = QMGR, and the -u argument is not set.

Action: Set the -u parameter, or change the queue, or queue manager configuration, to ANY.

Potential explanation: There are no active instances of the queue in the cluster.

Action: Check for instances of the queue with a CLWLPRTY value of 1, or greater.

Potential explanation: Remote queue instances have consumers ( IPPROCS >=1) but are inactive on those queue managers ( CLWLPRTY =0) because AMQSCLM is not monitoring those remote instances.

Action: Ensure that AMQSCLM is running on those queue managers, and / or the queue is in the list of queues to be monitored by checking the -q and -f parameter values.

Potential explanation: The remote queue instances are active ( CLWLPRTY =1), but are seen as inactive on the local queue manager ( CLWLPRTY =0). This situation is due to the updated CLWLPRTY value not being propagated to this queue manager.

Action: Ensure that the remote queue managers are connected to at least one of the full repository queue managers in the cluster. Ensure that the full repository queue managers are functioning correctly. Check that the channels between the full repository queue managers, and the monitored queue managers, are running.

Potential explanation: The messages are not committed, therefore they are not retrievable.

Action: Check that the sending application is functioning correctly.

Potential explanation: AMQSCLM does not have access to the local queue where messages are queued.

Action: Check whether AMQSCLM is running as a user with sufficient authorization to access the queue.

Potential explanation: The command server of the queue manager is not running.

Action: Start the command server of the queue manager.

Potential explanation: AMQSCLM encountered an error.

Action: Check the report files for errors.

Potential explanation: The remote queue instances are active (CLWLPRTY=1), but messages cannot be transferred to those queue instances because the cluster sender channel from the local queue manager is not running. This is often accompanied by a CLM0030W warning in the amqsclm report log.

Action: Check the status of the cluster sender channels from the local queue manager to the remote queue manager, or managers, with an active instance of the queue.