distributed host checks: freshness checking issues

[Pascal Vandeputte]

As good as Merlin sounds, we currently have something that works (including 
user-friendly managment & provisioning scripts) and unfortunately way too much 
other things on our plate to invest the appropriate amount of time in this.

But I'm certain that I will revisit these ideas in the future, so thanks a lot 
for all your suggestions! Much appreciated.

Best regards,

Pascal


On Tuesday 07 June 2011 15:35:55 Andreas Ericsson wrote:
> On 06/07/2011 02:28 PM, Pascal Vandeputte wrote:
> > On Tuesday 07 June 2011 10:12:25 Andreas Ericsson wrote:
> >> On 06/01/2011 05:51 PM, Pascal Vandeputte wrote:
> >>> Can anyone confirm that my reasoning is correct? That the master will
> >>> *always* keep on doing *some* host checks no matter what you configure?
> >> 
> >> More or less, yes. It will at least schedule them even if it gets
> >> results for them, but eventbroker modules can block even forced host
> >> checks. I'd look into using Merlin, DNX or mod_gearman if I were you.
> >> It will do what you want with far better performance than NSCA will
> >> ever be able to.
> > 
> > Thank you for the confirmation and the tips!
> > 
> > I've done some quick reading on Merlin, DNX and mod_gearman, and while
> > they look very interesting, our Nagios setup is probably a little too
> > complex to distribute the checks automatically the way we're doing it
> > now.
> 
> I doubt it. The largest setup running mod_gearman is checking something
> like 35000 services. The largest merlin install is running close to 48000.
> I'm sorry to say I have no figures for DNX.
> 
> > We have Nagios slaves ("workers") in multiple, independent locations, and
> > we're exploiting that to the fullest by running different service checks
> > of the same host from different locations: checks for "public" services
> > are effectively checked from a remote location, while NRPE checks to the
> > same host are done from within the same datacenter. The configs for each
> > master&  slave are generated from the same host/service database.
> 
> For this you'd have to use two host definitions if you use Merlin, since
> checks are split to workers based on hostgroup affiliation. Apart from
> that, everything should work just fine.
> 
> > Apparently, send_gearman can be used as a send_nsca replacement, which
> > I'll have to check up on later. It would be cool if it can send
> > multi-line output and performance data. NSCA cannot do multi-line and
> > doesn't seem 100% reliable either.
> 
> I'm fairly sure it can, but submitting passive checkresults with multiline
> output is not very well tested and may not work well with Nagios core. It
> will be particularly problematic if the check output is larger than 4096
> bytes, since that's the size of the command pipe. It works flawlessly with
> Merlin though, as does external command forwarding and a lot of other nifty
> things that NSCA-based setups simply doesn't handle.

------------------------------------------------------------------------------
EditLive Enterprise is the world's most technically advanced content
authoring tool. Experience the power of Track Changes, Inline Image
Editing and ensure content is compliant with Accessibility Checking.
http://p.sf.net/sfu/ephox-dev2dev
_______________________________________________
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