Monitoring / trending tools

[David Caro]

On 04/29, Ewoud Kohl van Wijngaarden wrote:
> Hello all,

> We all know the importance of monitoring and trending, but there's a lot
> of tools so I'd like to start the thread to look into those.

= Munin =
http://munin-monitoring.org/
Mostly meant for trending, though you can use the limits part to notify
on limits.

== Pros ==
* Easy to deploy
* Good puppet module available at https://github.com/duritong/puppet-munin

== Cons ==
* Hard to get custom graphs


= Nagios =
http://nagios.org/
Monitoring tool, with lots of plugins and very widely used.

== Pros ==
* Lots of plugins
* Lots of users
* Everyone has some knowledge
* Also has some puppet modules (even from puppetlabs)

== Cons ==
* Relatively complex to setup and maintain
* The ui is not as good as it could be


= Icinga =
http://icinga.org/
Nagios fork

== Pros ==
* All the nagios plus:
* Better ui
* Direct nagios configuration support (not sure if relevant here though)

== Cons ==
* Relatively complex to set up


= Zabbix =
http://www.zabbix.com/
Never used it, anyone?
Graphs and Monitoring

== Pros ==
* All in one

== Cons ==


= Ganglia =
http://ganglia.sourceforge.net/
Graphing tool for large clusters

== Pros ==
* Really scalable
* Comes with most used metrics by default (memory/cpu/...)

== Cons ==
* Not so easy to setup
* Difficult to generate custom graphs


= Graphite =
http://graphite.wikidot.com/
More Graphing

== Pros ==
* Easy to add new metrics
* Really nice interface (easy to create custom graphs, zooms,
* dasboards...)

== Cons ==
* Not so straight-forward to set up
* Might use more resources (dynamically generated graphs)


= pnp4nagios =
http://docs.pnp4nagios.org/pnp-0.6/start
Graphing plugin for nagios

== Pros ==
* Integrated with nagios, so it will use nagios/icinga checks for data
collection

== Cons ==
* Hard to make custom graphs


= Cacti =
http://www.cacti.net/
Graphing tool

== Pros ==
* Really configurable, has a lot of options (A LOT)
* That makes it hard to setup too

== Cons ==
Also rrd based, unable to create graphs dynamically


Should we open an etherpad with that info?

-- 
David Caro

Red Hat Czech s.r.o.
Continuous Integration Engineer - EMEA ENG Virtualization R&D

Tel.: +420 532 294 605
Email: dcaro at redhat.com
Web: www.cz.redhat.com
Red Hat Czech s.r.o., Purkyňova 99/71, 612 45, Brno, Czech Republic
RHT Global #: 82-62605