+

Search Tips   |   Advanced Search

Use basic or traditional message logs to troubleshoot applications

WebSphere Application Server can write system messages to several general purpose logs, including JVM, process, and IBM service logs, which can be examined for problem determination.

The JVM logs are created by redirecting the System.out and System.err streams of the JVM to independent log files. WAS writes formatted messages to the System.out stream. In addition, applications and other code can write to these streams using the print() and println() methods defined by the streams. Some Developer Kit built-ins such as the printStackTrace() method on the Throwable class can also write to these streams. Typically, the System.out log is used to monitor the health of the running application server. The System.out log and System.err log can be used for problem determination. The System.err log contains exception stack trace information that is useful when performing problem analysis.

Because each application server represents a JVM, there is one set of JVM logs for each application server and all of its applications located by default in the following directory:

In the case of a WAS ND configuration, JVM logs are also created for the deployment manager and each administrative agent because they also represent JVMs.

(ZOS) There is one set of STDOUT and STDERR log streams for each application server and all of its applications. JVM logs are also created for the deployment manager and each administrative agent because they also represent JVMs.

The process logs are created by redirecting the STDOUT and STDERR streams of the process to independent log files. Native code, including the JVM itself, writes to these files. As a general rule, WAS does not write to these files. However, these logs can contain information relating to problems in native code or diagnostic information written by the JVM.

As with JVM logs, there is a set of process logs for each application server, since each JVM is an operating system process. For WAS ND configuration, a set of process logs is created for the deployment manager and each administrative agent.

Deprecated feature: The IBM service log contains both the WAS messages that are written to the System.out stream and some special messages that contain extended service information that is normally not of interest, but can be important when analyzing problems. There is one service log for all WAS JVMs on a node, including all application servers. The IBM Service log is maintained in a binary format and requires a special tool to view. This viewer, the Log and Trace Analyzer, provides additional diagnostic capabilities. In addition, the binary format provides capabilities that are utilized by IBM support organizations.depfeat

In addition to these general purpose logs, WAS contains other specialized logs specific to a particular component or activity. For example, the HTTP server plug-in maintains a special log. Normally, these logs are not of interest, but we might be instructed to examine one or more of these logs while performing specific problem determination procedures. For details on how and when to view the plug-in log, see the Accessing a web resource through the application server and bypassing the HTTP server subsection of the A web resource does not display topic.

(Dist) Note: The system log (SYSLOG) is only supported on WAS for z/OS . WAS logging does not utilize operating system logs except in the case of z/OS.

(ZOS) Note: The System.out and STDOUT streams are redirected to the SYSPRINT ddname under z/OS. The System.err and STDERR streams are redirected to the SYSOUT ddname under z/OS. By default, the WAS for z/OS cataloged procedures associate these ddnames with print (SYSOUT=*) data sets, causing message logs to go into WAS job output. Job output can be viewed with the Spool Display and Search Facility (SDSF) or equivalent software.

Sometimes server and application problems can be diagnosed by examining log output from the WAS.


Tasks

Determine which type of logs we would like to implement:


Example

How to direct SYSPRINT and SYSOUT output to an HFS file.

If we are familiar with UNIX or Windows environments, we might be reluctant to use the facilities of SDSF (or IOF) to view the SYSPRINT and SYSOUT output from servants. If we would rather use a familiar editor (such as vi) in a Telnet session to view your output, it is possible to redirect the SYSPRINT and SYSOUT outputs to files in an HFS.

The following JCL example shows how to modify the SYSPRINT DD card in your startup procedure to redirect the output to an HFS file. The old SYSPRINT DD card has been commented out by preceding it with /*, and a new SYSPRINT DD card points to a file in the "/myDir/myServer" directory, in this case named was.log.d&LYYMMDD..t&LHHMMSS.log. The extra period between the date and time variables is not a typographical error, but rather an instance of JCL syntax necessary to terminate the first variable. &LYYMMDD will be replaced with the local date in YYMMDD format and &LHHMMSS will be replaced by the local time in HHMMSS format. The PATHMODE subparameter sets the file mode to 775 and the PATHOPTS subparameter OWRONLY opens the file for WRITE access. The sub-parameter OCREAT indicates that if the file does not already exist, create it.

We can modify the SYSPRINT DD card in either your Servant or Controller startup procedure. In addition, the SYSOUT DD card can be modified in the same way to redirect the SYSOUT output.

//*YSPRINT  DD SYSOUT=*,SPIN=UNALLOC,FREE=CLOSE
//SYSPRINT  DD PATHMODE=(SIRWXU,SIRWXG,SIROTH),
//   PATHOPTS=(OWRONLY,OCREAT),
//   PATH='/myDir/myServer/was.log.d&LYYMMDD..t&LHHMMSS'

If we try to direct the output for multiple streams to the same file, such as setting both DEFALTDD and HRDCPYDD variables, the allocation for the HRDCPYDD file fails and output is sent to the default location (JOBLOG/SYSLOG).


Subtopics

  • Web resource is not displayed