Chapter 9. System management 291
A major challenge of problem management is dealing with unanticipated issues. Much like
detective work, you need to find clues, make educated guesses, verify suspicions, and so on.
The most important skills are common sense, focus, thoroughness, and rigorous thinking. A
proactive approach to problem management is always the best choice. This section outlines
general practices to follow.
Perform the following checks to avoid issues with the runtime environment:
Check that you have the necessary prerequisite software up and running.
Check that the proper authorizations are in place.
Check for messages that signal potential problems. Look for warnings and error messages
in the following sources:
– Logs from other subsystems and products, such as TCP/IP, RACF, and Windows Event
– WebSphere Application Server SystemOut.log and SystemErr.log files
– SYSPRINT of WebSphere Application Server for z/OS
– Component trace output for the server
Check the ports used by WebSphere Application Server. The ports that WebSphere
Application Server uses must not be reserved by any other system component.
Check that enough disk space for dump files is available.
Check your general environment:
– System memory
– Heap size
– System has enough space for archive data sets
Make sure that all prerequisite fixes are installed. A quick check for a fix can save hours of
Become familiar with the problem determination tools that are available in WebSphere
Application Server and what these tools provide.
9.7.1 Log and traces
WebSphere Application Server V8 includes the following modes of logging:
High Performance Extensible Logging mode
High Performance Extensible Logging mode
Beginning in WebSphere Application Server Version 8, you can configure the server to use
the HPEL log and trace infrastructure instead of using SystemOut.log, SystemErr.log,
trace.log, and activity.log files or native z/OS logging facilities. By default, HPEL is not
enabled. You can enable it from the administrative console or by using wsadmin scripting.
Enabling HPEL: For instructions about enabling HPEL by using the administrative console
or wsadmin scripting, see the WebSphere Application Server Version 8 Information Center
at the following address, and search for basic mode and HPEL mode:
292 IBM WebSphere Application Server V8 Concepts, Planning, and Design Guide
After HPEL mode is enabled, the logs used in basic mode are no longer written to. HPEL
keeps log and trace data stored in a proprietary binary format in two repositories and a text
log file as illustrated in Figure 9-6:
Log data repository
The log data repository stores log records from applications or servers written to the
System.out, System.err, or java.util.logging file at the
Detail level or higher. Data
stored in the log data repository is most often useful to administrators.
Trace data repository
The trace data repository stores trace records from applications or servers that are written
to java.util.logging files at levels lower than
Detail. Data stored in the trace data
repository is most often useful to application programmers or by the WebSphere
Application Server support team.
The text log file content is redundant because the data in the text log file is also stored in
the log data and trace data repositories. The text log file is provided so that log content can
be read without using the LogViewer command-line tool to convert the log data repository
content to plain text. To improve server performance, the text log file can be disabled if the
LogViewer tool is always used.
Figure 9-6 HPEL mode content and routing
Log and trace performance is greatly enhanced with HPEL because of the following factors:
Log and trace events are stored only in one place.
Log and trace events are stored only in one place and not redundantly in several locations.
Log events, System.out, and System.err information is stored in the log data repository.
Trace events are stored in the trace data repository. If the text log file is disabled, data is
written only to these two repositories.