WebSphere Application Server V8.5 Administration and Configuration Guide for the Full Profile

620 WebSphere Application Server V8.5 Administration and Configuration Guide for the Full Profile

17.9 Logging and tracing

This section focuses on the logging and tracing options that are specific to the z/OS system

and the new standardized model of logging that is introduced in WebSphere Application

Server V8.5.

17.9.1 High Performance Extensible Logging overview

High Performance Extensible Logging (HPEL) is a complete solution for standardized logging

and tracing in WebSphere Application Server V8.5. It is an alternative to the existing log and

trace facilities that are offered on the z/OS platform, which use LogStreams, Component

Trace, Job Entry Subsystem, Hierarchical File System, or other facilities. It is a new strategic

logging solution delivered across platforms. It does not include native traces.

HPEL uses mechanisms with data repositories where logging and tracing data is stored all in

one place in binary format. To improve performance, data is not shared across instances and

is not converted by LogViewer to plain text unless needed. HPEL provides an optional text log

function for debugging convenience, where log or trace data can be written to text log file in

plain text format immediately. You can use the administrative console or the LogViewer

commands to view the HPEL collected logs and traces.

HPEL out performs existing log and trace facilities and can have performance benefits for

log-intensive applications; however, HPEL is not enabled by default. An HPEL API is provided

in the com.ibm.websphere.logging.hpel package to simplify development of tools for log and

trace processing.

For more information, go to the following website:

http://pic.dhe.ibm.com/infocenter/wasinfo/v8r5/index.jsp?topic=%2Fcom.ibm.webspher

e.zseries.doc%2Fae%2Fttrb_usinghpel.html

17.9.2 Enabling HPEL mode

To enable HPEL logging, refer to 19.4.5, “Advanced logging” on page 725. You must disable

text logging after enabling HPEL to improve performance.

17.9.3 z/OS logging and tracing tips

For WebSphere Application Server for z/OS in basic logging mode, the SystemOut.log,

trace.log, and STDOUT streams are redirected to the SYSPRINT ddname by default. The

System.err, and STDERR streams are redirected to the SYSOUT ddname. WebSphere

Application Server for z/OS cataloged procedures associate these ddnames with print

(SYSOUT=*) data sets, causing message logs to go into WebSphere Application Server job

output.

Chapter 17. Performance tuning 621

In basic mode, ensure that your logging and tracing level is not unnecessarily high by

completing the following steps:

1. Navigate to Troubleshooting  Logs and trace, as shown in Figure 17-15.

Figure 17-15 WebSphere Application Server Logs and trace view

2. Click the target application server name, as shown in Figure 17-16.

Figure 17-16 Application server choice on Logging and tracing view

3. Click Change log detail levels, as shown in Figure 17-5.

Figure 17-17 Change log detail levels on Application server Logging and tracing view

4. Verify your logging and trace detail level. The default logging level is *=info, as shown in

Figure 17-18 on page 622.

622 WebSphere Application Server V8.5 Administration and Configuration Guide for the Full Profile

Figure 17-18 Configuration view of the change log detail level view

For more information about trace controls:

http://pic.dhe.ibm.com/infocenter/wasinfo/v8r5/index.jsp?topic=%2Fcom.ibm.webspher

e.zseries.doc%2Fae%2Fwelc_ref_trb_file.html

In addition to the administrative console, you can also use WebSphere Application Server

V8.5 modify commands to change the tracing levels dynamically:

 F servername,TRACEALL={0,1,2,3}

Establishes a general trace level for the server.

 F servername,TRACEINIT

Resets the trace settings to the initial trace settings.

 F servername,TRACENONE

Turns off all trace settings.

 F servername,TRACETOSYSPRINT={YES|NO}

Selects whether to send the trace to SYSPRINT.

 F servername,TRACEBASIC=x and F servername,TRACEDETAIL=x

Sets the trace level, where x is a value from Table 17-2 on page 623 that represents a

component.

Preferred practice: Start the trace string from the most broad component groups and then

select more specific traces. The advantage to this approach is that the trace settings for

classes or packages that are contained in a larger group are specified correctly by

including them later in the trace string.

The logging string is processed from left to right. During the save processing optimization,

part of the logging string might be modified or removed if the levels they configure are

overridden by another part of the logging string.

Note: Subcomponents, specified by numbers, receive detailed traces. Other parts of the

product receive tracing as specified on the TRACEALL parameter.