Check intervals and notifications... retry_check_interval vs retry_interval

[Marc Powell]

On Sep 6, 2009, at 12:04 PM, Anthony wrote:

> I have a host definition with the following values:
>
> normal_check_interval 5
> retry_check_interval 1
> notification_interval 1
>
> Now, I receive the warning:
> Warning: Service ... on host ... has a notification interval less  
> than its check interval!  Notifications are only re-sent after  
> checks are made, so the effective notification interval will be that  
> of the check interval.

[snip]

> ... but this is noted in an example, not in the service definitions  
> section of the manual. On the other hand, in xodtemplate.c as  
> highlighted by someone else on this list in 2008, apparently they're  
> equivalent.

They are equivalent. retry_check_interval was renamed to  
retry_interval and is only valid for compatibility reasons. I expect  
it's going to be phased out entirely in a future release. They're  
currently treated exactly the same in the code.

>  two definitions of what retry_interval and retry_check_interval  
> would seem to be different. One suggests it's the time between  
> checks on a "non-OK" host. The other one suggests it's the time  
> between checks on a non hard/final state. Is it one, the other or  
> both?

It is used when the host/service is in a SOFT non-OK state. Once max  
check attempts is reached, the host/service is put in a HARD, non-OK  
state and nagios falls back to the normal interval for all further  
checks.

> Either way, would this not mean that if I had a normal check  
> interval of 5 minutes, and a retry check interval of 1 minute that  
> in some cases, it would indeed be checked more often than 5 minutes  
> and therefore be able to notify more often than the normal check  
> interval?

Nope. The notification interval is how frequently, after the first  
notification, additional notifications might be sent, bearing in mind  
that it's tied to checks. Since the first notification is sent out  
only after reaching a HARD state and nagios is only checking every  
normal_check_interval from that point forward, you can't notify less  
than normal_check_interval.

--
Marc


------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
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