Troubleshooting is the process of finding and eliminating the cause
of a problem. If you have a problem with your applications or your IBM software,
the troubleshooting process begins as soon as you ask yourself what happened?
Why and when to perform this task
If you encounter a problem with the functioning of service applications
or the server and bus environment, you can use this set of topics to help
you understand and resolve the cause of the problem.
Unlike performance
tuning, which focuses on solving problems associated with slow processes and
un-optimized performance, troubleshooting focuses on finding solutions to
functional problems.
This set of topics, and other aids for troubleshooting
with WebSphere ESB build
on information and other aids for troubleshooting with underlying WebSphere Application Server. Where
appropriate, links are made to the WebSphere Application Server information
and other aids.
As a general strategy for troubleshooting problems,
use the following steps:
Steps for this task
- Record any symptoms of the problem.
Depending on
the type of problem you have, whether it be with your application, your server,
or your tools, you might receive a message that indicates something is wrong.
Always record the error message that you see. As simple as this sounds, error
messages sometimes contain codes that might make more sense as you investigate
your problem further. You might also receive multiple error messages that
look similar but have subtle differences. By recording the details of each
one, you can learn more about where your problem exists.
- Look up information for any messages that you have.
If
you already have an error message and want to quickly look up its explanation
and recommended response, look up the message under Messages.
- Find any other messages that might help diagnose the problem.
For help in knowing where to find error and warning messages, interpreting
messages, and configuring log files, see Working with message logs.
- Try recreating the problem.
Think back to what steps
you were doing that led you to this problem. Try those steps again to see
if you can easily recreate this problem. If you have a consistently repeatable
test case, you have an easier time determining what solutions are necessary.
- How did you first notice the problem?
- Did you do anything different that made you notice the problem?
- If this worked before what has changed? The change can refer to any type
of change made to the system, ranging from adding new hardware or software,
to configuration changes to existing software.
- What was the first symptom of the problem you witnessed? Were there other
symptoms occurring around that point of time?
- Does the same problem occur elsewhere? Is only one machine experiencing
the problem or are multiple machines experiencing the same problem?
- What messages are being generated that could indicate what the problem
is?
- Eliminate known possible causes.
Narrow the scope
of your problem by eliminating components that are not causing the problem.
By using a process of elimination, you can simplify your problem and avoid
wasting time in areas that are not culprits. Consult the information in this
product and other available resources to help you with your elimination process.
Consider:
- Has anyone else experienced this problem?
- Is there a fix or a download?
- Investigate the product information for known problems based
on the task you were doing.
Depending on the specific problem
that you have encountered, it might be useful to troubleshoot based on different
issues that are known in installation, configuration, deployment, and administration.
Investigate known problems organized according to task areas in Troubleshooting by task.
- Examine Web-based resources for known problems and solutions.
In addition to the information center, you can examine a range of
Web-based resources for researching and resolving problems related to WebSphere ESB.
For more information about these resources, see Resources for diagnosing and fixing problems.
- Download and apply any fix for the problem.
A
product fix might be available to resolve your problem. You can determine
what fixes are available by running a query from the IBM Support Assistant.
For more information about getting and applying any fixes, see Getting fixes.
- If appropriate, take advantage of logging and common base events.
The support for logging and common base events is provided by the
underlying WebSphere Application Server.
- Applications can log diagnostic information while running. You can specify
how you want the server to handle log output, and what level of logging you
require. Using the administrative console, you can enable or disable a particular
log, specify where log files are stored and how many log files are kept, specify
the level of detail in a log, and specify a format for log output. You can
also set a log level for each logger.
Difficult problems might require the
use of tracing, which although conceptually similar to logging, exposes the
low-level flow of control and interactions between components. Collecting
a trace is often requested by IBM technical support personnel. If you are
not familiar with the internal structure of WebSphere Application Server,
the trace output might not be meaningful to you.
For general information
about using logging and tracing to diagnose problems, see Logging
and tracing with Java logging.
For more information about using
tracing, see Working with trace.
- An application can create an event object whenever something happens that
either should be recorded for later analysis or which might require additional
work to be triggered. Events are recorded with the standard structure called
the Common Base Event. A Common Base Event should contain all
of the information needed by the consumers to understand the event. This includes
information about the runtime environment, the business environment and the
instance of the application object that created the event. For more information
about using common base events for troubleshooting, see The
Common Base Event in WebSphere Application Server.
- Debug applications during development.
Before deploying
an application, assemble and test the application on a test server.
To
debug applications that run on WebSphere ESB,
you must use your application development tool. For more information, see "Debugging
components" in the WebSphere Integration Developer information center or
in the online documentation installed with IBM WebSphere Integration Developer.
The
IBM Developer Kit and Runtime Environment, Java 2 Technology Edition, Version
1.4.1 Diagnostics Guide describes debugging techniques and the diagnostic
tools that are available to help you solve problems with Java. It also gives
guidance on how to submit problems to IBM. You can find the guide at IBM developer kits: Diagnosis documentation Web site.
- Use WebSphere Application Server troubleshooting
capabilities
WebSphere ESB is
built on WebSphere Application Server Network Deployment. WebSphere ESB also
works with infrastructure and platform services from WebSphere Application Server.
For more information about troubleshooting in WebSphere Application Server,
select Troubleshooting
and support .
- If you cannot solve the problem, get help from IBM.
If
you are not able to resolve a problem by following the steps in this topic,
by looking up error messages in the message reference, or looking for related
documentation on the online help, contact IBM Technical Support.
To
find out how to look up documented problems, common mistakes, prduct prerequisites,
and other problem-determination information on the WebSphere ESB Web
site, or to obtain technical support from IBM, see Obtaining help from IBM.