Reporting for redundant hosts
Assaf Flatto
nagios at flatto.net
Thu May 20 11:32:21 CEST 2010
Stephan Tesch wrote:
> Hello everyone,
>
> I'm facing a problem for a long time now, having done some research on
> this, but never been able to find a solution:
>
> We're monitoring a lot of redundant network hosts (routers, firewalls,
> etc.) where we have to report the uptime of the whole setup, not only
> the uptime of a single component.
>
> An example:
>
> We have routerA and routerB, that form a HA setup. If one router is
> taken down, for example to update its firmware, or simply crashes, the
> whole setup still should have a reported uptime of 100%.
>
> What I've already tried is to configure a cluster service. The problem
> with this approach is, that I can't distinguish in the reporting whether
> the host was down or unreachable, which is another crucial number to
> report, also the downtimes don't get reflected very well.
>
> I'm running Nagios 3.2.1 with NDOutils if that matters for the question.
> Is there any plugin or addon that provides this functionality?
>
> Many thanks,
> Stephan
>
>
One approach is to monitor a VIP for the service (i.e. the "clustered"
ip of the external facing IP address of the routers) for finding if the
service is up , and the individual devices by their direct IP for each
device activity .
This will give you both the business view ( if the VIP is down then the
business is impacted) , and the operational view ( one device impacted ).
--
Never,Ever Cut A Deal With a Dragon
I am doing a Charity Bike ride On the 27 of June for the
Capital to Coast Charity. Please help by Donating
http://www.justgiving.com/Lovefilm-capital-to-coast
------------------------------------------------------------------------------
_______________________________________________
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
More information about the Users
mailing list