Nagios Support Forum

Hello. I'm looking for advice on designing a slightly complex monitoring environment with Nagios XI.

We have 2 data centers, Primary and Secondary. Each data center has its own internet connection. There is a dedicated private link between the data centers as well. The Primary data center uses 10.100.0.0/16 private IP addresses, and the Secondary 10.101.0.0/16. There is full network connectivity over the private link, so that a Primary host with an IP like 10.100.1.2 can connect to a Secondary host with an IP such as 10.101.3.4, and vice versa.

What i would like to achieve is to have Nagios running in each data center, monitoring the hosts that are local to it. I also want the Nagios instances monitoring each other. But i'd like configuration and alerting to be centralized to Nagios running in the Primary data center. The one exception is that if the Secondary Nagios detects that the Primary Nagios is down, it should send that alert directly.

Nagios Fusion was made for these types of scenarios! "Nagios Fusion functions as a high-level dashboard for multiple instances of Nagios Core and Nagios XI. Fusion acts as a central viewpoint for all servers, while leaving the actual monitoring to the individual Nagios instances."

Although, you don't need Fusion to accomplish your goals as having an instance of Nagios XI at each data center will suffice, and, these instances can monitor each other as well. But Fusion would make management thereof a lot easier.
Here is some documentation on the product, certainly let us know if you have any questions.

https://assets.nagios.com/downloads/nag ... erview.pdf
https://assets.nagios.com/downloads/nag ... tarted.pdf
https://library.nagios.com/library/prod ... giosfusion

Thanks for the feedback! I did look at Fusion a bit. It seems Fusion would provide a centralized dashboard but not centralized configuration. Of the common distributed monitoring solutions (DNX, Fusion, MNTOS) it looks like only DNX does centralized configuration, but with DNX it might be difficult to get the Secondary Nagios to alert independently if the Primary Nagios is down.

Maybe i'm overthinking the infrastructure and will just have to use 1 Nagios XI server that monitors everything in both data centers (that's what we have now). And then set up a second server that does nothing except monitor the first Nagios and alert if it goes down, but not try to tie the 2 together in any real way.

dramaley wrote:Maybe i'm overthinking the infrastructure and will just have to use 1 Nagios XI server that monitors everything in both data centers (that's what we have now). And then set up a second server that does nothing except monitor the first Nagios and alert if it goes down, but not try to tie the 2 together in any real way.

That's a very simple approach and one that is in-line with our licensing. Honestly unless you need two XI servers for network reachability reasons, 1 should suffice. I'd just say don't put more than about 20,000 hosts+services on a single machine - in that case you probably would want to split them between 2 servers.