Dependency issue

[John P. Rouillard]

In message <A61062A3CA04594180C8AAD1B06F6C25B27471 at kogex01.pmpl.co.uk>,
Deborah Martin writes:
>It seems that its very dependent on when the next scheduled check will be.
>If I set max_check_attempts to 5 for services
>B or C, and then have max_check_attempts set to 1 on service A, it makes no
>difference. Service B or C could still
>be scheduled to be checked before service A. In fact, on the first check
>attempt for service B, for example, a critical 
>alert is generated. 
>
>I want Nagios to detect that Service B or C is a dependency on Service A,
>and then if Service A is alerting, suppress alerts for Service B or C. 
>But nagios doesn't seem to do any kind of dependency lookup if a dependency
>is already alerting. Kind of defeats, in my view, the logic
>of dependencies. There is no point in Nagios telling me that Service B or C
>is down, when Service A is down. Service A must be UP for any of its
>dependencies to work. 
>
>If however, Service A is OK, then yes - Service B or C should alert if
>necessary. 
>
>If I were to have 20 dependencies against a master service, I don't want 20
>potential alerts being generated! Plus I want to know the EXACT cause of the
>problem not have to hunt it down every time. Isn't that one of the points of
>Nagios - to be able to finely tune plugin checks so that we can 
>easily investigate a problem ? It seems that I needn't bother with
>dependencies if this is how it's going to work. 
>
>Does anyone else agree / disagree because it seems to me that either the
>docs are not explaining dependencies correctly, or Nagios 
>is not behaving correctly. I would appreciate feedback on this issue. 

Try reading the nagios docs on soft_state_dependencies. Then setting

   soft_state_dependencies=1

in nagios.cfg.

If A depends on B (aka no alert on A when B in trouble) make sure that:

   retry_check_interval(A) * max_check_attempts(A) >
       normal_check_interval(B)

That should guarantee that at least one check of B occurs during the
trouble interval for A. Without the soft_state_dependencies you must use:

   retry_check_interval(A) * max_check_attempts(A) >
       normal_check_interval(B) +
       (retry_check_interval(B) * max_check_attempts(B))

to make sure that B goes into a hard state during the trouble interval
of A.

				-- rouilj
John Rouillard
===========================================================================
My employers don't acknowledge my existence much less my opinions.


-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click
_______________________________________________
Nagios-users mailing list
Nagios-users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. 
::: Messages without supporting info will risk being sent to /dev/null