nagios && rrd

[mark]

Sure. I looked at APAN and here is what I didn't like. The main thing 
was that I had to manually create an rrd store for each thing I wanted 
to graph. While this is fine for small, static shops, larger places with 
a somewhat dynamic base will quickly get tired of this. That is why I 
like Orca. All I have to do is feed it a log file and it creates and 
manages the rrd for me. Beyond install, I never have to touch rrd. The 
second, minor, thing is creating a service for each APAN item. Now it's 
true, I have to create a service for each thing I want to monitor 
anyway. But if I all ready have say 5k+ services, adding APAN to those 
could become tedious. Again, I'm thinking I could write some quick logic 
that feeds the data into Orca / rrd without my intervention. This should 
be easy as the HOSTNAME, HOSTALIAS, HOSTADDRESS, SERVICE, etc. are 
available. 

So, dynamic creation of rrd objects and a more variable based, centralized 
method for adding the APAN services would be my suggestions. Again, the 
dynamic rrd creation is the big one.

Mark

On Tue, 15 Apr 2003, Subhendu Ghosh wrote:

> Before you go off and start coding - any comments or suggestion for 
> improvement with APAN?
> 
> -sg
> 
> On Tue, 15 Apr 2003, mark wrote:
> 
> > I agree. This is a solution that I plan on implementing also. I have 
> > heard / seen at least one other person mention this also. Sounds like 
> > it's about time one (or all) of us got on with it and get Orca working 
> > with Nagios.
> > 
> > 
> > Jason, did you have any ideas on how you were going to hand the data to 
> > Orca? I see two options, you could run a process for each piece of data, 
> > which would be each event(?), and run Orca in the rrd insert only mode. 
> > Or, you could write to a log file and have Orca batch load the data. The 
> > second is my choice but I haven't come up with a clear / clean way to 
> > write the data to logs in a format that Orca likes. I think I would need 
> > a log for each host and possibly for each service. Anyone else come to a 
> > similar conclusion?
> > 
> > Mark
> > 
> > 
> > On Tue, 15 Apr 2003, John Sellens wrote:
> > 
> > > | From: Jason Burnett <jason at cannonfodder.org>
> > > | Date: Thu, 20 Mar 2003 17:15:55 -0500
> > > | 
> > > | Is anyone else using RRD to process the service-perf data? If so, what
> > > | are you using to parse out the $PERFDATA$ & $OUTPUT$ info? We are
> > > | working on perl to parse it out and put it in RRD for us, but if
> > > | someone has found an easier way to do this please let me know.
> > > 
> > > My belief, which I haven't implemented yet, is that a good
> > > graphing/trending mechanism is to log the host and service perfdata
> > > from within Nagios, and then use Orca (http://www.orcaware.com/orca/)
> > > to insert the data into RRD files and manage the HTML interface.
> > > 
> > > I'm reasonably optimistic that this is a good idea.
> > > 
> > > John
> > > jsellens at generalconcepts.com
> > > 
> 



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
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