Some people have the need to check more often than every minute so they've changed the check_interval to a fraction (0.5 means 30 seconds, 0.25 means 15 seconds).
Icinga Server: OS Linux-2.6.18-238.12.1.el5-x86_64 (CPUs: 8 x 2833 Mhz), 32189 Mb mem
[RedHat EL5 on a Sun x4150]
Icinga 1.6.1
gearmand 0.25
mod-gearman 1.1.0
livestatus 1.1.12p7
pnp4nagios 0.6.16
All of the above are from source.
~ 1200 hosts, ~8000 services. Classic Console only.
We have a request to change one or two checks so they run every 30 seconds, and eventually every 15. This is to match our original (internally built) monitoring environment - The process check we use there was modified to make it compatible with Icinga/NRPE, but the default interval_length is too long for us.
Anyone have any experience (good/bad/other) in changing the interval_length below 60? I'm already certain that we will need to split off a dedicated checking host for it, if we keep this Active. Our program can usually finish within 2-3 seconds on a local host (after checking every process), so that's not a problem.
I'm also going through tests that would convert these to Passive/NSCA, but we need to re-compile that first to display larger output. We may still need the lowered interval however, for alerting/auto-restart when there's no recent update.
Thanks,
--TSK
=====-----=====
Tim Kimball
http://sungak.net
------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:
Build for Windows Store.
http://p.sf.net/sfu/windows-dev2dev
_______________________________________________
icinga-users mailing list
icinga-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/icinga-users
Gesendet: Freitag, 21. Juni 2013 um 12:45 Uhr
Von: tim.kimb...@sungard.com
An: icinga-users@lists.sourceforge.net
Betreff: [icinga-users] Running a check every X Seconds
Von: tim.kimb...@sungard.com
An: icinga-users@lists.sourceforge.net
Betreff: [icinga-users] Running a check every X Seconds
Icinga Server: OS Linux-2.6.18-238.12.1.el5-x86_64 (CPUs: 8 x 2833 Mhz), 32189 Mb mem
[RedHat EL5 on a Sun x4150]
Icinga 1.6.1
gearmand 0.25
mod-gearman 1.1.0
livestatus 1.1.12p7
pnp4nagios 0.6.16
All of the above are from source.
~ 1200 hosts, ~8000 services. Classic Console only.
We have a request to change one or two checks so they run every 30 seconds, and eventually every 15. This is to match our original (internally built) monitoring environment - The process check we use there was modified to make it compatible with Icinga/NRPE, but the default interval_length is too long for us.
Anyone have any experience (good/bad/other) in changing the interval_length below 60? I'm already certain that we will need to split off a dedicated checking host for it, if we keep this Active. Our program can usually finish within 2-3 seconds on a local host (after checking every process), so that's not a problem.
I'm also going through tests that would convert these to Passive/NSCA, but we need to re-compile that first to display larger output. We may still need the lowered interval however, for alerting/auto-restart when there's no recent update.
Thanks,
--TSK
=====-----=====
Tim Kimball
http://sungak.net
------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:
Build for Windows Store.
http://p.sf.net/sfu/windows-dev2dev
_______________________________________________
icinga-users mailing list
icinga-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/icinga-users
------------------------------------------------------------------------------ This SF.net email is sponsored by Windows:
Build for Windows Store. http://p.sf.net/sfu/windows-dev2dev
_______________________________________________ icinga-users mailing list icinga-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/icinga-users
