Nagios Support Forum

Hi all,

I am right now in the final part to deploy a Nagios XI solution with the following architecture using openstack:

DC1 (production):
2 servers as Nagios XI frontends
3 servers as BBDD (MySQL Enterprise with innoDB)

DC2 (PRE/DEV/TEST - some productive environments):
2 servers as Nagios XI frontends
3 servers as BBDD (MySQL Enterprise with innoDB)

I don't know which would be the best solution in this case, replicate DDBB, configuration files and so on or, federate both systems, the main requirement is that in case that for any reason the Nagios XI instance on DC1 felts down, DC2 Nagios instance must be able to continues providing Monitoring services, even for all the systems that are in DC1 and viceversa, from my point of view federated would be better but I don't know if there is an issue to cross all information between both systems, under my understanding both should work as "master"

Any help is welcome.

BR

Hi Danzur,

There are a few ways to achieve a high availability or similar solution with Nagios XI. A simple method that many customer employ is to maintain backups of the primary server, and import those into a secondary server in the event of a failure (see first doc in the list below). Another option is to setup a HA cluster for Nagios XI on Linbit.

The following documents can help with a setup like this:
https://assets.nagios.com/downloads/nag ... ios-XI.pdf
https://assets.nagios.com/downloads/nag ... ios-XI.pdf
https://www.linbit.com/en/576-ha-nagios ... -on-rhel7/

Thanks for the reply,

I have been checking the documentation, I think database replication is not a good solution because lets say both are somehow master, I mean, both will monitor different systems, if i do a database replication I will delete information in one of them.

The front End of the servers and the DDBB are already in HA so i don't know if would be a good idea also create the HA between the two sites, but, I understand that federating both Nagios the information will be shared between both and each one can assume the monitoring if for any reason one of them fails.

is it right?

BR

Nagios XI has the ability to send/receive transfer checks from other XI servers, and this is something used in distributed environments, but not intended to create high availability architecture since configurations are local to each server.

For more information, we have several knowledgebase articles for configuring inbound/outbound checks.
https://assets.nagios.com/downloads/nag ... ith_XI.pdf
https://assets.nagios.com/downloads/nag ... ith_XI.pdf

Hello back,

Finally I decided to replicated the database between the two CPDs and keep the DR servers switched off.

Thanks for your help and support.

BR

Hello @danzur,

Great. Thanks for getting back to us. Are we ok to close this topic?