Chapter 4. Tools for determining performance problems 81
4.2.7 HEAPANA command
In 4.2.6, “iDoctor” on page 40, you learned about iDoctor and how you can use its tools, such
as Job Watcher and Heap Analysis Tools for Java, to gain an in depth understanding of your
application. IBM also supplies an operating system command-line interface with iDoctor that
helps manage the collection of data for heap-related issues. The command is called
HEAPANA.
HEAPANA handles the execution of the Table Dumps, Profiling, and Job Watcher collections
described previously in 4.2.6, “iDoctor” on page 40. It provides a single interface, which
simplifies your collection activities. In addition, HEAPANA ensures that the data you collect
using the various iDoctor tools captures the exact same time period. This enables you to
move easily between data derived from different tools as you research performance
problems.
Usage
The HEAPANA command is shipped with Job Watcher. To execute the HEAPANA command,
you need to add the QPYRTJW library to the library list and simply type HEAPANA on an iSeries
command line. Press F4 to view the parameters for the command, as illustrated in
Figure 4-70.
Figure 4-70 HEAPANA: Executing the HEAPANA command
The available parameters for HEAPANA are explained here. To see all parameters, press F9.
JVM Job Name, JVM Job User, JVM Job Number: The qualified job name of the JVM.
Run as Trigger?: This parameter indicates whether you intend to use Job Watcher trigger
support to monitor the heap and then trigger the collection of additional information. For
details on the Job Watch trigger support, refer to “Job Watcher trigger support” on
page 85.
Library to save data to: This is the library where the collected data is saved. You can
name it as you choose. If it doesn’t exist, it is created.
Interval: This parameter is used to determine how often snapshots of the heap are taken.
The “right” value for this parameter depends on a few factors, such as how long you intend
the collection to run and how rapid the growth is. You need to ensure that enough
snapshots are taken so that you can have a feel for the growth that is occurring.
Samples: This indicates the number of snapshots that will be collected.
Heap Analysis. (HEAPANA)
Type choices, press Enter.
JVM Job Name . . . . . . . . . . > SERVER1
JVM Job User . . . . . . . . . . > QEJBSVR
JVM Job Number . . . . . . . . . > 012243
Run as Trigger ? . . . . . . . . YES
Library to save data to . . . . > ITSCIDXXHA
Interval(secs between samples) > 60
Samples (# of collections) . . . > 30
Output Queue . . . . . . . . . . PAGESTATS
Job Watcher Intervals to get . . > 10
Profiler Timer (Seconds) . . . . > 30

Get Maximum Performance with WebSphere Application Server V5.1 on iSeries now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.