Resolving trace problems when developing message flows

Follow this advice to deal with some common trace problems that can arise when you develop message flows.

About this task

You cannot determine which node is being referenced in your trace file

Procedure

  • Scenario: You have generated a service trace file for your message flow, to trace the path of a message. However, you cannot determine which node is being referenced in the trace file.
  • Explanation: In the trace file you might see text such as:
    Video_Test#FCMComposite_1_1 ComIbmMQInputNode , Video_Test.VIDEO_XML_IN
    The elements in the text have the following meanings:
    Video_Test
    is the name of the message flow
    FCMComposite_1_1
    is the internal name for the node
    ComIbmMQInputNode
    is the type of node
    VIDEO_XML_IN
    is the node label, and is the name you see in your flow
    The number at the end of the internal name is incremented; for example, FCMComposite_1_4 would be the fourth node you added to your flow. In the example, this section of the trace is referring to the first node in the message flow.

You cannot see any alerts when you change user trace

Procedure

  • Scenario: You cannot see any alerts for an integration server or message flow in the Alerts viewer when you use the mqsichangetrace command to change the user trace setting.
  • Explanation: No alert is generated when an integration server runs user trace at normal or debug level. Alerts are generated only for message flow trace. For message flow trace, the IBM® Integration Toolkit is not notified of trace changes that are initiated by the mqsichangetrace command.
  • Solution: To see the alert, refresh the message flow, or disconnect, then reconnect to the domain.

Data that the Trace node sends to the syslog on UNIX is truncated

Procedure

  • Scenario: You are using a Trace node on UNIX and have set the Destination property to Local Error Log. You send a message to the Trace node that consists of multiple lines, but the message that appears in the syslog is truncated at the end of the first line.
  • Explanation: On UNIX, syslog entries are restricted in length and messages that are sent to the syslog are truncated by the new line character.
  • Solution: To record a large amount of data in a log on UNIX, set the Destination property on the Trace node to File or User Trace instead of Local Error Log.