Skip to content

IsmaelKP2/Nagios

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 

Repository files navigation

Nagios

ITSI correlation searches for Nagios

Documentation for the Add-on here : https://docs.splunk.com/Documentation/AddOns/latest/NagiosCore/About

This add-on must be installed on a Linux instance of Splunk Enterprise for data collection. The add-on is platform independent for indexers and search heads.

ITSI versions tested for this package : https://splunkbase.splunk.com/app/1841/ Documentation for ITSI https://docs.splunk.com/Documentation/ITSI/latest

What is Nagios where to download it https://www.nagios.org/projects/nagios-core/

What is a correlation Searches how to configure it : https://docs.splunk.com/Documentation/ITSI/4.3.1/Configure/Correlationsearchoverview

  • How To -

#First install the Nagios Add-on :

https://splunkbase.splunk.com/app/2703/

Verify that you have an eventgen.conf files and a samples folder

Please find the samples find in /opt/splunk/etc/apps/Splunk_TA_nagios_core/samples/

service-perfdata.samples for service performance data nagios_notifications.samples for Alerts data

#Second install EventGen :

http://splunk.github.io/eventgen/SETUP.html#install

make sure to enable the EventGen both in Manage Apps -> SA-EventGen and in Settings->Data Inputs->SA-EventGen

Restart your Instance

Verify that the data is ingested by running the following search index=* spourcetype=nagios*

service_name service_shortname are two fields that shows the different KPIs

#Correlations Searches

you can find the correlations package in this Github repo

ItsiDefaultScheduledBackup-1573486962.zip


If you want to build your own

here is the search :

Index=“YOUR_NAGIOS_SERVICEPERFDATA_INDEX” | eval norm_severity=case(  service_status==“CRITICAL”,6,
  service_status==“WARNING”,4,  service_status==“OK”,2)| dedup consecutive=true host_name service_status service_name| eval tmp_entity=host_name | eval sec_grp= “default_itsi_security_group” | ‘match_entities(tmp_entity,sec_group)’| ‘filter_maintenance_entities’

First Line I do an eval to map the severity to the severity in ITSI The critical service status severity maps to The ITSI norm severity of six The third line here is done to do a dedup of the event. So that actually you don't get duplicates we do it here on hostname service status and service name. So, provided that you have those in your data it is going to work. Fourth Line tmp_entity is the entity we have in ITSI we map it here to the host_name

The last lines you don’t need to worry it's added automatically to the search

We want now to build the aggregations policy to group the event there is one by default

but let us use the severity and the service_name for the grouping to get good enough results.

About

ITSI correlation searches for Nagios

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors