Page 1 of 1
Nagios Distributed Monitoring
Posted: Thu Aug 06, 2015 3:31 am
by lafargeuser
I have went through Nagio XI distributed monitoring.There are methods to distributed monitoring.
Fusion
DNX
suppose we have 1500+ devices & 20000 service metrics spead across all locations. Here what can be good approach to implement,bit confused what to use when.
Re: Nagios Distributed Monitoring
Posted: Thu Aug 06, 2015 10:44 am
by jdalrymple
DNX is (currently) a dead project, don't use it.
mod_gearman is our currently supported way of distributing the load of a single Nagios server to multiple worker servers, this could be a good option.
Fusion doesn't distribute the load, it aggregates the view of multiple Nagios installations into 1 centralized single pane of glass. There is no central configuration in this architecture.
The other one you didn't mention, and my favorite is to have multiple remote Nagios servers send passive checks using obsessive commands back to a single server. This also doesn't have a central configuration, but it does allow for massive scalability without the need for add-on products.
Whichever direction you choose, you can't go wrong. It all depends on your situation which option would be the best. Here is our official documentation on the topic:
https://assets.nagios.com/downloads/gen ... utions.pdf
Re: Nagios Distributed Monitoring
Posted: Fri Aug 07, 2015 1:37 am
by lafargeuser
Thanks for brought to my notice.
and my favorite is to have multiple remote Nagios servers send passive checks using obsessive commands back to a single server. This also doesn't have a central configuration, but it does allow for massive scalability without the need for add-on products.
Can you brief this bit more that how do I achive this multiple remote Nagios server send passive checks. Are you refering Mod Gearman ?
Re: Nagios Distributed Monitoring
Posted: Fri Aug 07, 2015 10:47 am
by jdalrymple
I did a brief write up on it here:
https://support.nagios.com/forum/viewto ... nd#p140693
The basic concept is that remote Nagios servers run checks, then for every check result (good/bad/other) they run a command, usually something like send_these_results_to_my_central_nagios_server.sh
The central nagios server does no active checking, it just sits and collects passive checks from the remote servers.
Again - no 3rd party involvement.