Setup a redundant Nagios server. (Offsite)

[QS1]

Any pointers on setting up another nagios server just as a back up? I think we are allowed to have one setup in the event the production server goes down. Would the backup server be able to automatically start up as the production in the event it goes offline. Also, will the checks and configuration automatically be copied over to the backup nagios server?

Any documentation on this will be helpful.
Thanks!

CentOs release 5.4
32 bit
VMware Image running on ESX
GNOME 2.16.0

[QS1]

edit:
http://nagios.sourceforge.net/docs/2_0/redundancy.html

I saw this, will this work, if so how about the XI portion?

[tonyyarusso]

The Core documentation will not be sufficient for this, no, and I'm afraid we haven't tested this sort of thing to write up documentation specific to XI. To do what you want, you would need to set up replication with both MySQL and PostgreSQL databases, as well as syncing the Nagios flat files and status information. You'd then need to custom-script starting monitoring services on the backup if the production server went down.

In short, there would be a fair bit of work involved to accomplish this, and we simply haven't gotten around to figuring out what would be necessary yet.

[QS1]

Ok, well any suggestions on doing anything close to this? I was under the impression we atleast got the backup server free?

[mguthrie]

I'm not sure what the licensing covers for a redundant setup, that's a question that you'd have to ask our sales team ([email protected]).

Here's a document for backing up and restoring an XI install.
http://library.nagios.com/library/produ ... -nagios-xi
You could potentially have this processes scripted to sync the configs and databases between the two servers. Unfortunately we don't have a syncing process like this documented, but maybe this will point you in the direction you're looking at.

We've discussed having an enterprise addon that setup a process like this for you, and maintained synchronization between the two servers, and the second server was set to activate the moment it detected the 1st server went down. Any feedback or thoughts on this idea?

[QS1]

We've discussed having an enterprise addon that setup a process like this for you, and maintained synchronization between the two servers, and the second server was set to activate the moment it detected the 1st server went down. Any feedback or thoughts on this idea?

Yeah that's exactly the way we wanted it setup. That would be great !!!

[mguthrie]

Well that might be a ways off before we have that ready. The variation on how people want this setup can be pretty huge, but we appreciate the feedback!