Note: The trace string you put in depends on what module
in Application Server you need to trace or what problem you are wishing to
troubleshoot. In order to determine what string you need to put in, please
consult the WebSphere
Support site . Once there, search on "MustGather and" place in the
module you wish to troubleshoot. This will narrow down your search for the
particular MustGather document. Once you get to the MustGather document,
follow the instructions within to get the trace string.
For V6.0:
Static Trace:
1. Log on to the administrative console.
2. In left panel expand Troubleshooting. Click on Logs and
Trace.
3. Select the application server to be traced, and than on the next
page click the Diagnostic Trace link.
4. Select the Configuration tab.
5. Select the Enable Log property.
6. Under Trace Output, select File radio button and
specify File Name. Also Increase the Maximum file size to
100 MB and Increase the Maximum number of historical files to
10.
7. Select Basic (Compatible) Trace Output Format unless
otherwise specified by support.
8. Navigate to Logging and Tracing >
your_application_server > Change Log Detail Levels.
9. Under Configuration tab specify Trace Specification by
entering the trace string specific to the mustgather doc for which you are
collecting data.
10. Click Apply and OK. Then Save your
configuration (select Synchronize changes with Nodes option).
Dynamic Trace:
1. Log on to the administrative console.
2. In the left panel, expand Troubleshooting and click on
Logs and Trace.
3. Select the application server to be traced, and than on the next
page click the Diagnostic Trace link.
4. Selet the Runtime tab.
5. Under Trace Output, select Memory Buffer and type a
Dump file name (if you don't specify path but just file name than
default location of file is under application server profile directory).
Set the Maximum buffer size to 300. This equates to approximately
60 MB of additional Java™ heap usage. Make sure your application server
JVM Heap Max value is set appropriately to accommodate the additional heap
usage.
Important: Do not select Save Runtime Changes to Configuration as
well if you do not want this setting to become permanent.
6. Navigate back to Logging and Tracing >
your_application_server > Change Log Detail Levels.
7. Select the Runtime tab.
8. Under Trace Specification, enter trace string for the
specific mustgather for which you are collecting data.
9. Click Apply, then OK to activate the trace.
10. Reproduce the problem.
11. To dump the traced data to a file, navigate to Logging and
Tracing > your_application_server > Diagnostic Trace
Service.
12. In the Runtime tab click Dump to generate the trace
output to the specified Dump File.
For V5.0 and V5.1:
Statice Trace:
1. Log on to the administrative console.
2. In left panel expand Troubleshooting. Click on Logs and
Trace.
3. In right panel click on your_application_server >
Diagnostic Trace.
4. In Configuration tab, check Enable trace.
5. In Trace Specification write trace string from the mustgather
doc for which you want to collect data.
6. In the Trace Output, choose File Name and type a file
name. Increase the Maximum file size to 100 MB. Increase Maximum
number of historical files to 10 as well.
7. Select Basic (Compatible) Trace Output Format unless
otherwise specified by support.
8. Click Apply then OK and than Save your
configuration (select Synchronize changes with Nodes option).
Dynamic Trace
1. With the application server started, start the administrative
console.
2. In left hand panel, expand Troubleshooting and click Logs
and Trace.
3. Select the application server to be traced, and then on the next
page click the Diagnostic Trace link.
4. Select the Runtime tab.
5. Under Trace specification, write trace string specific to
mustgather you are collecting.
6. In the Trace Output panel, select the Memory Buffer
radio button.
7. Enter 200 in the Maxmum Buffer Size field. This equates to
approximately 40 MB of additional Java™ heap usage. Make sure your
application server JVM Heap Max value is set appropriately to accommodate
the additional heap usage.
8. Click apply and OK. Trace is now activated.
9. Reproduce the problem.
10. To dump the traced data to a file, return to the Logs and
Traces page and select the Runtime tab of Diagnostic Trace
Service.
11. Specify a Dump File Name (for example:
logs/myTraceData.log).
12. Click Dump to generate the trace output to the specified
Dump File Name.
Application Server Developer provides a Test Environment. Using this
Test Environment, you can run Application Server traces that will assist
you in troubleshoot problems.
To enable a WebSphere Application Server trace:
- Open the server configuration editor from the Server Perspective of
WebSphere Studio Application Developer.
- Select the Trace tab. From here, the Trace string and Trace output
file can be configured.
- Save your changes.
The next time your test server starts, tracing is enabled. In the
WebSphere Studio Application Developer console, the following output is
displayed:
TRAS0017I: The startup trace state is <listing of the trace string.
The trace output is sent to the specified Trace output file.
You can also enable tracing in the WebSphere Administrative Console if
the administrative console is enabled on the test server.
To enable the administrative console:
- Select the Configuration tab in the server configuration editor.
- Select the Enable administration console checkbox.
- Restart the Test Server.
- Open a Web browser and point to http://localhost:9090/admin to access
the administrative console.
- Enable tracing in the same way that it is enabled when running outside
of WebSphere Studio Application Developer.
|