IBM Integration Bus provides several tracing methods to troubleshoot the problems occurring in message flows applications. You could trace the message flows, integration servers, commands and the integration nodes. Based on the depth of tracing enabled, the trace could be user trace or service trace.
The detailed concepts are explained here
Until IBM Integration Bus V9.0.0.3, once the trace is enabled, the tracing continues until it is stopped explicitly by another command. This sometimes led to over tracing after the problem is captured which made reading the trace cumbersome.
This post talks about the “Trigger tracing” feature included in IBM Integration Bus V9.0.0.3. With this feature, tracing can be enabled with a trigger, in the form of a BIP message, on the occurrence of which tracing will be disabled automatically. This feature is only applicable to integration server service trace.
The details of the feature are listed here:
1. The integration server service tracing can be enabled with up to 10 BIP message numbers as trigger and the tracing stops on the first occurrence of any one those BIP numbers.
Example command:
mqsichangetrace IIBV9NODE –e default –t –l debug –m temp=2208,BIP2112
With the above command using option “-m temp=2208,BIP2112”, trace is enabled with the trigger to stop the tracing on occurrence of either BIP2208 or BIP2112.
The BIP message used as trigger needs to be a number with/without “BIP” prefixed. The severity of the number i.e., I/E/S/W is optional. Also, the BIP message is case insensitive.
If you need to disable the trace before the trigger is hit then the tracing can disabled normally.
When the tracing is enabled with a trigger and if the trigger is not hit at all, the tracing gets disabled on integration server/integration node restart.
2. The mqsireporttrace command output reflects the BIP numbers used as trigger to disable trace.
Example command:
mqsireporttrace IIBV9NODE -e default -t
BIP8098I: Trace level: debug, mode: temp=2208,2112, size: 102400 KB.
BIP8071I: Successful command completion.
3. Notifications are available in eventlog whenever trace is enabled or disabled either manually or when the trigger is hit.
BIP2297I : ( IIBV9NODE.default ) Integration Server service trace has been enabled due to user-initiated action.
BIP2296I : (IIBV9NODE.default ) Integration Server service trace has been disabled due to user-initiated action.
BIP2223I : (IIBV9NODE.default ) Integration Server service trace has been enabled with trigger to stop the trace on occurrence of any of the BIP messages : ”2208”
BIP2221I : (IIBV9NODE.default ) Integration Server service trace has been disabled due to the occurrence of BIP message ‘2208’ which was specified as a trigger to disable tracing.
4. If an integration server is continuously restarting, then its “single restart” can be captured easily using the below steps:
a. Stop the integration node
b. Enable trace : mqsichangetrace IIBNODE –e default –t –l debug –m temp
The trace is automatically disabled if the integration server restarts after broker startup.
5. Similarly, integration server startup can be traced and stopped on hitting a trigger. “mqsichangetrace” with “-m temp=