Disabling Freshness Checks? Service dependencies under freshness?

[Mueller, Karl]

Hello all,

I recently got our distributed Nagios environment up and running.  Cool
stuff.

We're running into a problem, though.  Under the previous active checks
setup, you can run the "DISABLE_HOST_SVC_CHECKS" and "DISABLE_
HOST_CHECK" commands to turn off checks of hosts and services.  We use
this when we're doing a short maintenance, for example.  Scheduling
downtime is actually more work than just turning it off and on via a
simple script.

In the distributed environment, the slave servers correctly stop
performing active checks.  However, the master server also does not
receive service result updates on those services.  As a result, it
triggers active service checks when their freshness expires.  These, of
course, are forced, despite active checks being enabled or not.  That
triggers some escalations when the forced service check fails. 

I have looked through the source code and it appears that the freshness
checking ignores almost anything else going on.   I can think of two
possible solutions:   First, disable notifications for the hosts in
addition to active checks.  While this won't prevent active checks from
being run, at least stuff won't be sent out about it. 

The fact that active checks even occur could be bad, though.  There may
be handlers and such which may run.  All this leads me to believe that
there should be a way to turn off the freshness checking.  I believe
that it would be possible to create an external command which can turn
on/off the "check_freshness" variable, similar to the
"DISABLE_HOST_SVC_CHECKS", we could have "DISABLE_SVC_FRESHNESS_CHECK"
or such.  (Along with the 'enable' portion)

Also, on a more serious note, it looks like freshness checking ignores
service dependencies.  This of course means that when any slave server
stops checking something due to a service dependency, the master server
goes stale and runs active checks.  Is this intentional?

I'd be willing to write some patches for these issues, unless there are
concerns.

Thanks for any input,
Karl




-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id149&alloc_id66&op=click