Nagios Support Forum

Our goal: to continue monitoring in a Disaster Recovery scenario, by failing critical production servers over to DR location.

The Window's and Linux SA have done this DR drill successfully before, but not with Nagios monitoring.

I believe the Hostname will be the same, however subnet will be different.

What "gotcha's" do we need to watch out for regarding Nagios? (i.e. is licensing tied to the IP?)

Note: I have host groups setup for Prod & DR so we can set scheduled downtime for Prod after failing over to DR.

Environment: Offline (Nagios server may not have internet access)
VM
RHEL 7.8
Nagios XI 5.7.4

I'd also like to use this solution sync Staging with Production, so we can test/validate upgrades in Staging prior to Prod.

I suspect someone else has already done a similar activity and kept notes on what worked, and what didn't, so please share!

Thanks,

Craig

Hi MonitorGuy,
I searched our forum and found this post/ticket, which is related to your topic.

https://support.nagios.com/forum/viewto ... 16&t=49570

Regards,
Vinh

Here's the internal response I've received for this task:

Linux SA:

What we want to do is clone the PROD Nagios server (which is all up-to-date), and transfer that clone to Staging, and blow it onto the Nagios server there, so everything matches between environments.
 
To make that transfer happen, we need to mount a LUN (temporarily) from one environment in the other environment.
 
------

Windows SA:

Once this process is approved, here’s a high level of what needs to happen;
 
· Clone the Linux production Nagios server and note which LUN the cloned VM is on.
 
- To test the new LUN when it was added, I had moved the Windows production vCenter server onto it. 
 
· Engage the SAN team to temporarily present the production LUN that the cloned VM is on, to the non-production cluster.
· Once the LUN can be seen by the non-production cluster, move the cloned VM off of the production LUN and onto a non-production LUN.
· Then the SAN team would then need to un-present the production LUN from the non-production cluster.
· The cloned VM will now have been moved from the production LUN to a non-production LUN and ready for whatever configurations are needed.
 
------

SAN Team:

From the SAN OPS perspective, presenting an existing LUN to another group of hosts (and then taking it away from those hosts later), is about as “no risk” an activity as there is.

------

Besides changes needed on the Linux VM in staging (hostname, IP, etc.) what changes might Nagios require in Staging to pull off this clone?

I'll be using staging to validate upgrades, and test new monitor solutions prior to changing production.

Thanks!

Hi,
To pull this off and bring your DR server up and running, you might want to make sure your current Nagios XI license is valid for your new servers.

Please contact our sale team directly at "[email protected]" regarding that.

Regarding remote server like NRPE, you might also want to add the new DR's IP to its config.

For examples NRPE's agent:
/etc/xinetd.d/nrpe


Regards,
Vinh