[Nagiosplug-help] Output of check_http not consistent

[Thomas Guyot-Sionnest]

On 23/05/07 02:26 AM, Ingo Lantschner wrote:
> Am 23.05.2007 um 07:35 schrieb Thomas Guyot-Sionnest:
>> Here's what we do in a perfdata caching daemon we use (Old version at:
>> http://control-alt-del.org/code/NPDaemon/ ; I guess now is a good time
>> to post the updated version to NagiosExchange... Should be there  
>> soon.).
> Interesting! Thanks for the hint. Just one question: What happens, if  
> the polling server gets down - does the perfdata get cached on the  
> Nagios host or is it lost until the polling server gets up again?

Even though we can run Cacti on a different server (connection uses TCP)
we're currently using the same for graphing (just as we did with MTRG
and a few other RRD-based stuff before Cacti)

So right now the data is discarded after 2 minutes and overridden as it
comes in... That because our services are checked every minute and Cacti
is patched (and modified because their 1-minute patch is incomplete*) to
poll every minutes. I will eventually add command-line options or at
least some variables at the top :)

* IMHO broken since to really support multiple graphing times you need a
lot of duplicates RRAs and data templates. I suggested a smart way of
fixing it some time ago but they never replied.

I don't know any polling system that can fetch data in the past after a
downtime period so anyways I doubt this would be useful. If you really
want you could rewrite NPDaemon so that it writes itself the RRDs, but
you'll need a graphing system that allow to easily create graphs from
external RRDs.

BTW the updated NPDaemon code if on NagiosExchange (I just uploaded it):

http://www.nagiosexchange.org/Misc.36.0.html?&tx_netnagext_pi1[p_view]=972

Thomas