Escalate after X warnings or criticals

[Andrew Li]

Does anyone know if the notification count problem got fixed in 3.2.1?

I had a read of the ChangeLog but it doesn't mention anything related to
this problem since 3.0.6.

Cheers,
Andrew


On Mon, 2009-11-09 at 08:55, Neil Ramsay wrote:
> Hi Martin,
> 
> The escalation_options don't take the state into consideration during
> the notification count. So if you have an escalate rule on the 4th
> notification and only escalate on Critical in the escalation_options
> then following scenario is can occur: 
> You have 3 warning notifications and the 4th is Critical then it will
> escalate as there have been 4 notifications and a Critical. I posted a
> help request on this issue a week or two ago and would really like
> this to be patched or built into the next update.
> http://article.gmane.org/gmane.network.nagios.user/64997/match=escalation+state
> 
> Cheers,
> 
> Neil
> 
> On Sat, Nov 7, 2009 at 12:56 AM, Martin Melin <mmelin at gmail.com>
> wrote:
>         The existing escalation_options directive in escalation
>         definitions will likely get you this behavior without the need
>         for a patch.
>         
>         http://nagios.sourceforge.net/docs/3_0/escalations.html - see
>         the very bottom of this page as well as the object definition
>         documentation for escalation_options.
>         
>         Regards,
>         Martin Melin
>         
>         On Fri, Nov 6, 2009 at 3:49 AM, Mark Gius
>         <mgius at createspace.com> wrote:
>                 Currently, service notifications contain
>                 "first/last_notification"
>                 directives, that specify the range of notifications
>                 that the escalation
>                 should apply to.  This method of escalation has a
>                 weakness however.
>                 
>                 At my work, we let warnings go to the default contact
>                 (which happens to
>                 be email), and escalate to a pager chain on critical.
>                 However, if a
>                 service sits in WARNING for a length of time (which is
>                 likely to happen
>                 in the middle of the night), by the time the service
>                 enters a CRITICAL
>                 state the notification count exceeds our highest
>                 escalation, and our
>                 entire team gets paged immediately.
>                 
>                 What I'd like to see is the ability to distinguish
>                 between a WARNING
>                 notification and a CRITICAL notification in the
>                 escalation, and set up
>                 escalation chains that work based on the number of
>                 CRITICAL's that have
>                 been sent, as opposed to the total number of
>                 notifications.
>                 
>                 I am planning on patching nagios to support this
>                 behavior if there isn't
>                 a way to achieve this behavior with the current
>                 implementation.  My plan
>                 is to add a warning/critical count to service, add a
>                 first/last
>                 warning/critical state to service escalations, and add
>                 the directives
>                 "(first|last)_(warning|critical)_notification" to the
>                 service escalation
>                 configs.  The idea is also to keep the current
>                 behavior
>                 (notification_count and first/last_notification would
>                 still be present),
>                 but allow finer grained control over when escalations
>                 are sent out.
>                 This way if somebody didn't want to use the finer
>                 grained control their
>                 behavior would stay the same.  My current plan is to
>                 match the
>                 escalation if _any_ of the 3 notification ranges match
>                 (all/warning/critical).
>                 
>                 Any advice on making this behavior happen with Nagios
>                 as-is, or
>                 suggestions/advice on the implementation are welcome.
>                 
>                 -Gius
>                 


------------------------------------------------------------------------------
ThinkGeek and WIRED's GeekDad team up for the Ultimate 
GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
lucky parental unit.  See the prize list and enter to win: 
http://p.sf.net/sfu/thinkgeek-promo
_______________________________________________
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