AW: Distributed checking

Verstege, Bernhard Bernhard.Verstege at hik.fzk.de
Thu Feb 6 13:10:48 CET 2003


Hi,
works as designed ...
if a service check returns a non ok state (active or passive), a host check
is performed.
The other thing is the enabled feshness-check. If the distibuted monitor
does not send a passive check for the given threshold, an active check
on your master will be performed.
These checks will be performed, even if active checks are disabled !

To make sure the distributed monitor is still alive, we use the
check_dummy plugin. We use two different service definitions:
- the distributed monitor does a check_dummy every 5 minutes for example
  and sends the result (allways OK) to the master
- on the master we defined a freshness threshold of 10 minutes and
  a "check_command" which does no real check. It only sends an email 
  to the admin which tells him that the distributed monitor has not
  send an "alive" check for the given threshold.
- all other checks on the master have freshness_check disabled

Hope this helps !

Gruß 
B. Verstege
 
-------------------------------------------------------------------------
Bernhard Verstege
Forschungszentrum Karlsruhe GmbH Abt. HIK
Technik und Umwelt
Hermann-von-Helmholtz-Platz 1         Tel.: 0049 7247-82-5665
D-76344 Eggenstein-Leopoldshafen      Fax.: 0049 7247-82-4972
email:  verstege at hik.fzk.de




>-----Ursprüngliche Nachricht-----
>Von: Jason Burnett [mailto:jason at cannonfodder.org]
>Gesendet: Mittwoch, 5. Februar 2003 20:01
>An: nagios-users
>Betreff: [Nagios-users] Distributed checking
>
>
>o.k. I am using the distributed method with passive checks to monitor
>some hosts I have behind a firewall that I dont want to punch holes in.
>For some reason my central nagios server keeps running active checks on
>the hosts even though I have that disabled on a per service 
>instance. So
>I end up with the service up when it gets the info via the 
>passive tests
>then it goes critical when the central server tries to check it with an
>active test. Here is the service config from the central server:
>
>define service{
>    host_name           pinky
>    service_description     SSH
>    check_command           check_ssh
>    max_check_attempts      2
>    normal_check_interval       5
>    retry_check_interval        5
>    passive_checks_enabled      1
>    active_checks_enabled       0
>    check_period            24x7
>    flap_detection_enabled      1
>    process_perf_data       1
>    retain_status_information   1
>    retain_nonstatus_information    1
>    notification_interval       1
>    notification_period     24x7
>    notification_options        w,u,c,r
>    notifications_enabled       1
>    check_freshness         1
>    freshness_threshold     180
>    contact_groups          admins
>}
>
>
>-- 
>


-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com




More information about the Users mailing list