On Fri, 9 Aug 2013, st...@linuxsuite.org wrote:

> I am considering icinga for monitoring and performance graphing.

    Some level of caution is advised here.  Whilst it is easy to
confuse monitoring/alerting and trending the two are actually
different animals.  Icinga is superb as a monitoring/alerting
suite; trending and graphing, depending on *precisely* what you're
after may be better handled outside of Icinga (e.g. Cacti).

    The above having been said, addons to Icinga (e.g. nagiosgraph)
exist that will generate time-series graphs from the performance
data returned from the various plugins but are, of necessity, limited
to what the plugins return.

    Writing a custom plugin to monitor, say, traffic on a router or
on a particular is easy enough, and can return the number of inbound
and outbound octets or packets for a single interface when can then
be interpreted by, say, nagiosgraph to produce your plots.

    Monitoring the health state of a host, however, is quite a bit
more complex and requires the examination of many factors (memory
useage (both physical and virtual), load average, CPU usage, and
response-time to name a few) and combining those into a mathematical
model of how healthy the host is from a capacity perspective.

    I wrote a plugin a number of years to do such analyses, and it
worked OK for the generation of graphs of all the various parameters,
but suffered from scheduler stalls when Nagios (which is what I was
using at the time, this being "before the fork") would repopulate the
database following a restart and would produce graphs with gaps in
them.  This offended the author and I rewrote it to be called from
cron and pass the data for the monitoring application in via passive
checks and pass raw data off to a secondary routine that populated
the RRDs behind the scenes.  This got me clean, unbroken, graphs and
much more control over the presentation than nagiosgraph can provide.
(This may have been fixed with the most recent version of Icinga
which uses a proper FIFO when writing to the database so the actual
check-scheduler lights off quite quickly upon (re)start.)

    So, depending on your exact needs, off-the-shelf stuff may work
for you or you may have to invest a little bit of time in learning
some programming and write your own.

    Cheers!

+------------------------------------------------+---------------------+
| Carl Richard Friend (UNIX Sysadmin)            | West Boylston       |
| Minicomputer Collector / Enthusiast            | Massachusetts, USA  |
| mailto:crfri...@rcn.com                        +---------------------+
| http://users.rcn.com/crfriend/museum           | ICBM: 42:22N 71:47W |
+------------------------------------------------+---------------------+

------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite!
It's a free troubleshooting tool designed for production.
Get down to code-level detail for bottlenecks, with <2% overhead. 
Download for free and get started troubleshooting in minutes. 
http://pubads.g.doubleclick.net/gampad/clk?id=48897031&iu=/4140/ostg.clktrk
_______________________________________________
icinga-users mailing list
icinga-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/icinga-users

Reply via email to