Nagios Support Forum

We have a 3.2.1 Core installation on a Centos 5.5 server. I don't mind shutting the existing Core down, as it's not used for production monitoring at this point, but want to install XI on this server without doing a fresh Centos install.

My plan was to rename /usr/local/nagios -> /usr/local/nagiosOLD then proceed with your "Manual Installation Instructions".

The only other evidence of the old Core installation that I can find is:
- /etc/httpd/conf.d/nagios.conf exists
- startup files in /etc/init.d, rc3.d, and rc5.d exist
- nagios user, group, and empty /home/nagios exist
- /etc/selinux/targeted/modules/active/modules/nagios.pp exists

I suppose I could delete or rename all these too.

How does my install plan sound? Any other advice?

Thanks....Lyle

It's fairly likely that you'll hit some snags along the road if you do it that way, however it can be done if your system needs require it. Our "recommended" manual install is for a clean system.

Some things to note if you're wanting to install this way:
-The manual install scripts install Nagios core, XI runs on top of the Nagios core engine.
-If you're not using mysql or postgresql, I would uninstall them if you can and let the scripts re-install them and setup the users for it. Otherwise the root user needs to have the password "nagiosxi" for both DB's.

-You can import a Core configuration into Nagios XI. This is a common practice, and I would recommend installing XI on a clean system and doing a config import rather than the way you're suggesting. I would take a look a look at the following doc if you're going to use this method.
http://assets.nagios.com/downloads/nagi ... ngfromcore