Introduction¶
NMON is short for Nigel’s Performance Monitor, it is available on AIX Systems, Solaris (with Sarmon), and Linux Systems.
nmon for Linux is open source under GPL: http://nmon.sourceforge.net
nmon for AIX is not open source but is integrated into topas command from:
- AIX 5.3 TL09
- AIX 6.1 TL02
See: http://www.ibm.com/developerworks/wikis/display/WikiPtype/nmon
- nmon for Solaris (formerly sarmon) is open source and available for Solaris 10/11: http://www.geckotechnology.com/sarmon
This is a great “all in one” Performance Monitor tool that provides a very large amount of system performance information, and it can be used in different scenarios.
The first way to use NMON is running the “nmon” command in terminal, which opens a Real time monitoring terminal interface, giving you access to many system metrics within a single screen:
nmon is also very often used as a Capacity Planning and Performance tool by running the nmon process in a csv generating mode all along it’s run time, for later cold Analyse.
Here are some useful links about NMON:
- http://nmon.sourceforge.net/pmwiki.php
- http://www.ibm.com/developerworks/aix/library/au-analyze_aix
- http://www-01.ibm.com/support/knowledgecenter/ssw_aix_71/com.ibm.aix.cmds4/nmon.htm
- http://www-01.ibm.com/support/knowledgecenter/ssw_aix_71/com.ibm.aix.cmds5/topas.htm
- http://nmon.sourceforge.net/pmwiki.php
- http://www.geckotechnology.com/fr/sarmon
Analysing NMON csv data is not easy because it has a very specific format Splunk cannot directly manage. (one big problem stands in the event time stamp identification which is very uncommon and defined by a non time stamp pattern)
This is why I decided to develop this App, based on my own professional experience in Unix systems Capacity Planning, to provide to anyone interested a powerful too to Analyse NMON data with an Enterprise Class Application.
How it works¶
In a few words, here is how the App works:
- The Nmon core application contains all the views, data models, configurations and related objects
- The TA-nmon which is the technical addon for the Nmon Performance application contains binaries and scripts to manage the nmon data
- The TA-nmon once installed starts immediately to collect and transforms nmon performance and configuration data
- The default configuration indexes data into the “nmon” index (by default)
Splunk Data structure¶
nmon_data¶
Performance metrics data ordered by the key “type” which corresponds to the nmon section metric item (CPU_ALL, LPAR…):
index=nmon sourcetype=nmon_data
Eventtype::
eventtype=nmon:performance
nmon_config¶
Configuration data extracted by nmon2csv converters, corresponds to AAA and BBB* sections of nmon raw data:
index=nmon sourcetype=nmon_config
Eventtype::
eventtype=nmon:config
nmon_collect¶
Output of the nmon_helper.sh script which is responsible for nmon instances launches:
index=nmon sourcetype=nmon_collect
Eventtype::
eventtype=nmon:collect
nmon_processing¶
Output of nmon2csv Python and Perl converters (conversion of nmon raw data into csv data):
index=nmon sourcetype=nmon_processing
Eventtype::
eventtype=nmon:collect
nmon_clean¶
Output of the nmon_cleaner.sh script (interface to nmon_helper.py | nmon_helper.pl) which is responsible for nmon raw data file cleaning:
index=nmon sourcetype=nmon_clean
Eventtype::
eventtype=nmon:clean
Available packages¶
There are different packages:
- The *Nmon core* Application: this is the whole package you download in Splunk App (directory called “nmon”)
- The PA-nmon_light addon, available in the Git repository https://github.com/guilhemmarchand/PA-nmon_light (tgz archive), this package is expected to be deployed in indexers
- The TA-nmon addon, available in Splunk base https://splunkbase.splunk.com/app/3248 and https://github.com/guilhemmarchand/TA-nmon (tgz archive), can be deployed to any AIX / Linux / Solaris Full or Universal forwarder instance, master node of a cluster, deployment server, standalone instances, clustered indexers…