https://github.com/mnalis/nagios-local-multi

nagios report multiple problems determined by existance of files
https://github.com/mnalis/nagios-local-multi

nagios-plugin

Last synced: about 16 hours ago
JSON representation

nagios report multiple problems determined by existance of files

• Host: GitHub
• URL: https://github.com/mnalis/nagios-local-multi
• Owner: mnalis
• License: gpl-3.0
• Created: 2017-09-02T15:17:36.000Z (about 7 years ago)
• Default Branch: master
• Last Pushed: 2023-12-31T01:20:35.000Z (11 months ago)
• Last Synced: 2023-12-31T02:26:36.415Z (11 months ago)
• Topics: nagios-plugin
• Language: Perl
• Homepage:
• Size: 52.7 KB
• Stars: 0
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
This nagios check is intended to be run via NRPE on client computer

It will check /var/spool/nagios-local-multi/ for any visible non-zero-length files

containing any output not starting with "OK:", and report them as errors to nagios.

Regexp for OK/CRITICAL can be defined per-file basis in /etc/nagios/local-multi.conf

If file has "timestamp" in the name, it will also be checked that it was updated in last 48 hours.

Files ending in ".tmp" or starting with "." are ignored.

Examples for use:

- simply redirect your cron output to file /var/spool/nagios-local-multi/cron

  and you'll get nagios alerts instead of emails

- avoid creating a bunch of checks on nagios server for every little thing:

  just make a quick /etc/cron.daily/something script which writes error output 

  to /var/spool/nagios-local-multi/something (it can be as simple as adding at the 

  top of shell script with "set -e; exec 2> /var/spool/nagios-local-multi/something")

- some scripts are included by default as example (like debian apt(8) check and cron staleness check)

- example /etc/cron.daily/zimbra_running

  sudo -u zimbra zmcontrol status | grep -v Running > /var/spool/nagios-local-multi/zimbra_running

- example /etc/cron.hourly/ganeti_sanity

  gnt-cluster verify | egrep 'ERROR|WARNING|NOTICE' > /var/spool/nagios-local-multi/ganeti_sanity

(Note: All files must be readable by user which nagios/NRPE runs as, so do an 

umask 000 first in your scripts if needed)

Best feature: you only add the check via NRPE once to your nagios server, and all other scripts are only 

quickly created on client and automatically report to nagios.

This is best used a safety net for scripts which "should never fail" -- if your script expect to raise 

nagios alerts fairly often, it is probably better to dedicate separate nagios check for them, as 

nagios output is limited, and nagios-local-multi will combine all error output together to send to NRPE.

License: GPLv3 or higher, Matija Nalis , started 2017-09-02