Determining application delivery failure of messages
This scenario describes how you can use the WebSphere® Message Broker Monitoring agent to determine the cause of the problem when messages that are destined for an application are not being received by that application.
Perform the following steps to enable the WebSphere Message Broker Monitoring agent to gather the appropriate statistics:
- Place one CandleMonitor node immediately after the MQInput node so that the entry of the messages into the message flow is recorded.
- Set the type attribute of this CandleMonitor node to input.
- Place another CandleMonitor node immediately in front of the MQOutput node so that the exit of the messages from the message flow is recorded.
- Set the type attribute of this CandleMonitor node to output.
- Start the WebSphere Message Broker Monitoring agent.
After you have placed and set the CandleMonitor nodes, and the WebSphere Message Broker Monitoring agent is running, perform the following procedure:
- Open the Message Flow Status workspace and examine the Status attribute. The Status column for the monitored message flow must have the value Started. If it does not, start the message flow manually.
- Open the Monitor Node Message Flow Statistics workspace and compare the Current Message Input Count and Current Message Output Count attributes. A discrepancy between these values indicates that messages are being lost in the monitored message flow.
- Open the Monitor Node Events workspace and determine whether any exceptions have been triggered by a node in the monitored message flow.