include:
- Perl
- PHP
- MySQL
- Jpgraph
For each service type you wanted to track, you would create a separate table
(eg, disk space, CPU, whatever). A perl script would munge the data from
the performance log, and stuff the output into the respective tables. (A
future release of Nagios may make this step even simpler.)
To view a graph, clicking the relevant link would hit the PHP code, which
would slurp the appropriate column for the appropriate table, with options
for basing the output on all the data, or some arbitrary interval (eg, every
5th row). Some more PHP code would feed the output into Jpgraph and create
a graph on-the-fly.
Advantages of this approach vs. using RRDtool (as I see it):
- you can keep data as long as you like; you're not limited by the
predefined RRDs at create time
- you're writing the data to one database once; you're not writing it
multiple times (typically 3 times) for separate RRDs
- you only read from the database when you need a graph generated; a graph
is only generated when you wish to see it, and not on a constant basis
- the database can be on another (centralized, database) server
- you can selectively remove data (eg, purge old rows prior to a certain
date)
- I suspect that this will scale better (in terms of performance)
Just my $0.02.
jc
> -----Original Message-----
> From: Edwin Eefting [mailto:lists@datux.nl]
> Sent: Monday, January 20, 2003 6:00 AM
> To: Nagios-devel@lists.sourceforge.net
> Subject: [Nagios-devel] performance data gathering?
>
>
> I'm currently thinking of a new way to do performance data
> gathering with
> nagios and rrdtool. (to create those nice graphs
>
> Currently there's a nice project called apan, but it doesn't
> intergrate
> nicely with nagios. (you'll have to do a some things double,
> configure a
> lot of complicated stuff etc.)
>
> I purpose the following method to add performance data graphs
> to nagios:
>
> A generic way to let the current plugins return performance data that
> nagios can store in rrdtool. This way it wouldn't matter if i use
> check_disk with nrpe, or nagios_statd, or any other method.
> The current
> plugins only need a slight modification to return the data in
> some generic
> format. Perhaps we can add a "field" to the current returned
> data and let
> plugins return something like:
>
> DISK OK - [1000000 kB (10%) free on /dev/sda1]
> curr=9000000,max=10000000
>
> So we add a new line with performance data thats easy to
> interpret and is
> formatted in a generic way. So "curr" is always the current value of
> a service, and "max" always will be the maximum value.
>
> What do you think of this? Any suggestions for this system,
> or should we
> just do it and submit our patches?
>
> --
>
> //||\\ Edwin Eefting
> || || || DatuX - Linux solutions and innovations
> \\||// http://www.datux.nl
>
>
>
> -------------------------------------------------------
> This SF.NET email is sponsored by: FREE SSL Guide from Thawte
> are you planning your Web Server Security? Click here to get a FREE
> Thawte SSL guide and find the answers to all your SSL
> security issues.
> http://ads.sourceforge.net/cgi-bin/redi ... thaw0026en
> _______________________________________________
> Nagios-devel mailing list
> Nagios-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/lis ... gios-devel
>
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: jcarro10@sprintspectrum.com