Question & Answer
Question
How to collect perfpmr on AIX 7.1
Cause
Performance issues found in AIX 7.1 . Diagnostic data requested for problem resolution.
Answer
This perfpmr package contains a number of performance tools and some instructions. Some of these tools are products available with AIX. Some of the tools are prototype internal tools (setpri, setsched, iomon, getevars, pmucount, lsc, fcstat2, memfill, getdate, perfstat_trigger, kheap, tcpstat, foldstat, tstat, pvpa) and are not generally available to customers. Anyone who downloads or uses these tools must agree to these terms:
These tools are not officially supported and are provided "AS IS".
In addition, tools such as pmucount and pmucount.sh should not be used unless directed to do so by IBM level 3 performance
support.
NO WARRANTY FOR THE PROGRAM, THE PROGRAM'S RESULTS, OR INFORMATION CONTAINED IN THE PROGRAM
You are responsible for the results obtained from the use of the Program. Any results of the Program are provided "AS IS".
All results generated by the Program are estimates and averages based on certain assumptions and conditions. Each environment has its own unique set of requirements that no tool can entirely account for. No representation is made that the results will be accurate or achieved in any given IBM installation environment. The result is based on specific configurations and run time environments. Customer results will vary. Any configuration recommended by the Program should be tested and verified. Any code provided is for illustrative purposes only.
IBM does not warrant the accuracy, currency or completeness of information contained in the Program, including, but not limited to any information provided to IBM by any third party, and undertakes no obligation to correct or update such information. Any reliance by You on the Program or any results from the Program, including any information from any third party, is at Your sole risk and will not create any liability or obligation for IBM. IBM provides no assurances that any reported problems may be resolved with the use of any information that IBM provides.
Further, IBM does not control the transfer of data over the Internet. IBM does not warrant secure operation of the Program. In addition, IBM does not warrant that it will be able to prevent third party disruptions of the Program.
---------------------------------------------------------
AIX 7.1 PERFORMANCE DATA COLLECTION PROCESS
---------------------------------------------------------
Note: The act of collecting performance data will
add load on the system. HACMP users may
want to extend the Dead Man Switch timeout
or shutdown HACMP prior to collecting perfpmr data
to avoid accidental failovers.
TABLE OF CONTENTS
-----------------
I. INTRODUCTION
II. HOW TO OBTAIN AND INSTALL THE TOOLS ON AN IBM RISC SYSTEM/6000.
III. HOW TO COLLECT DATA FOR AN AIX PERFORMANCE PROBLEM
IV. HOW TO SEND DATA TO IBM
I. INTRODUCTION
This package contains a set of tools and instructions for
collecting the data needed to analyze a AIX performance
problem. This tool set runs on AIX V7.1
II. HOW TO OBTAIN AND INSTALL THE TOOLS ON AN IBM RISC SYSTEM/6000.
A. OBTAINING THE PACKAGE
The package will be distributed as a compressed
"tar" file available electronically.
From the internet:
==================
'ftp://ftp.software.ibm.com/aix/tools/perftools/perfpmr'
B. INSTALLING THE PACKAGE
The following assumes the tar file is in /tmp and named
'perf71.tar.Z'.
a. login as root or use the 'su' command to obtain root
authority
b. create perf71 directory and move to that directory (this
example assumes the directory built is under /tmp)
# mkdir /tmp/perf71
# cd /tmp/perf71
c. extract the shell scripts out of the compressed
tar file:
# zcat /tmp/perf71.tar.Z | tar -xvf -
III. HOW TO COLLECT DATA FOR AN AIX PERFORMANCE PROBLEM
A. Purpose:
1. This section describes the set of steps that should be
followed to collect performance data.
2. The goal is to collect a good base of information that
can be used by AIX technical support specialists or
development lab programmers to get started in analyzing
and solving the performance problem. This process may
need to be repeated after analysis of the initial set
of data is completed.
B. Collection of the Performance Data on Your System
1. Detailed System Performance Data:
Detailed performance data is required to analyze and
solve a performance problem. Follow these steps to
invoke the supplied shell scripts:
NOTE: You must have root user authority when executing
these shell scripts.
a. Create a data collection directory and 'cd' into this
directory.
Allow at least 45MB*#of_logicalcpus of unused space in
whatever file system is used.
*IMPORTANT* - DO NOT COLLECT DATA IN A REMOTELY MOUNTED
FILESYSTEM SINCE IPTRACE MAY HANG
For example using /tmp filesystem:
# mkdir /tmp/perfdata
# cd /tmp/perfdata
b. HACMP users:
Generaly recommend HACMP deadman switch interval
be lengthened while performance data is being
collected.
c. Collect our 'standard' PERF71 data for 600 seconds
(600 seconds = 10 minutes). Start the data collection
while the problem is already occurring with the command:
/directory_where_perfpmrscripts_are_installed/perfpmr.sh
The perfpmr.sh shell provided will:
- immediately collect a 5 second trace (trace.sh 5)
- collect 600 seconds of general system performance
data (monitor.sh 600).
- collect hardware and software configuration
information (config.sh).
In addition, if it finds the following programs available
in the current execution path, it will:
- collect 10 seconds of iptrace information (iptrace.sh 10)
- collect 10 seconds of filemon information (filemon.sh 10)
- collect 60 seconds of tprof information (tprof.sh 60)
NOTE:
Since performance problems may mask other
problems, it is not uncommon to fix one issue and
then collect more data to work on another issue.
d. Answer the questions in the text file called
'PROBLEM.INFO' in the data collection directory
created above. This background information
about your problem helps us better understand
what is going wrong.
IV. HOW TO SEND THE DATA TO IBM.
A. Combine all the collected data into a single binary 'tar' file
and compress it:
Put the completed PROBLEM.INFO in the same directory
where the data was collected (ie. /tmp/perfdata in
the following example). Change to the parent
directory, and use the tar command as follows:
Either use: cd /tmp; perfpmr.sh -o perfdata -z pmr#.pax.gz
or
# cd /tmp/perfdata (or whatever directory used
to collect the data)
# cd ..
# pax -xpax -vw perfdata | gzip -c > pmr#.pax.gz
B. Submission of testcase to IBM:
Internet 'ftp' access:
----------------------
The quickest method to get the data analyzed is
for the customer to ftp the data directly to IBM.
Data placed on the server listed below cannot be
accessed by unauthorized personnel.
Please contact your IBM representative for the
PMR#, BRANCH#, and COUNTRY#. IBM uses all 3 to
uniquely associate your data with your
problem tracking record.
'ftp testcase.software.ibm.com'
Userid: anonymous
password: your_internet_email_address
(ie. smith@austin.ibm.com)
'cd toibm/aix'
'bin'
'put PMR#.BRANCH#.COUNTRY#.pax.gz'
(ie. '16443.060.000.pax.gz'
'quit'
If the transfer fails with an error, it's possible that a file already exists by the same name
on the ftp server. In this case, add something to the name of the file to differentiate it from the file
already on the ftp site (ex. 16443.060.000.july18.pax.gz).
Notify your IBM customer representative you have
submitted the data. They will then update the
defect report to indicate the data is available
for analysis.
This data will then be transferred electronically
to AIX Development by 'ftp'.
IBM Software Service in the geography should inform
the Austin AIX Support Center when the performance
data has been sent to Austin by updating the PMR.
'ftp' address:
testcase.software.ibm.com
(use the procedure for internet ftp access described above)
Related Information
Was this topic helpful?
Document Information
Modified date:
15 September 2021
UID
isg3T1026140