Nagios Support Forum

Do you guys have any documentation on possible DR setups for Nagios XI.

This DR will be a 1 to 1 equivalent cold site. We would like to keep both PROD & DR in sync via automation.
To complicate matters we have 3 instances using PostgreSQL and 1 instance using MySQL only.

Please advice on the best method to go about syncing the following
1. Configurations
2. RRD files
3. Databases
4. status.dat

Thanks

This document has some initial suggestions:
https://assets.nagios.com/downloads/nag ... ios-XI.pdf

And there was a presentation at a previous Nagios conference by a former support tech on this topic:
https://www.youtube.com/watch?v=KW5Qkl8brcA

LINBIT offers some technical docs that might be of use if DRBD is a tool you're interested in leveraging:
https://www.linbit.com/en/576-ha-nagios ... -on-rhel7/

The documentation is a bit thin though. Built-in high availability is something that is on the Nagios Core 5 roadmap, which I imagine will eventually make it's way into XI in some form or another.

Thanks for the links, @mcapra!

mcapra wrote:This document has some initial suggestions:
https://assets.nagios.com/downloads/nag ... ios-XI.pdf

And there was a presentation at a previous Nagios conference by a former support tech on this topic:
https://www.youtube.com/watch?v=KW5Qkl8brcA

LINBIT offers some technical docs that might be of use if DRBD is a tool you're interested in leveraging:
https://www.linbit.com/en/576-ha-nagios ... -on-rhel7/

The documentation is a bit thin though. Built-in high availability is something that is on the Nagios Core 5 roadmap, which I imagine will eventually make it's way into XI in some form or another.

Thanks. Most of it is more of a HA setup where else my setup is at 2 different locations.
Anyone else has any other inputs?

I use a couple scripts to sync my live systems to standby systems. The main script runs from cron so the standby can be 5 minutes (or whatever the cron runs) older, but the NXI DB and configuration is pretty static, I don't see it as an issue. This takes care of numbers 1 and 3 in your list. Status.dat and RRD files can be synced w/ rcopy, but as RRD files change frequently, there might be an issue there.
The script takes a backup of the DB on the primary, restores it on the standby and then applies the config to write config files.
The standby has monitoring disabled until the primary fails and failover is manual... The license agreement says a secondary system shouldn't be actively monitoring or something like that. The cron on the primary would need to be disabled when a failure occurs.

Thanks for the input, @gormank!

gormank wrote:I use a couple scripts to sync my live systems to standby systems. The main script runs from cron so the standby can be 5 minutes (or whatever the cron runs) older, but the NXI DB and configuration is pretty static, I don't see it as an issue. This takes care of numbers 1 and 3 in your list. Status.dat and RRD files can be synced w/ rcopy, but as RRD files change frequently, there might be an issue there.
The script takes a backup of the DB on the primary, restores it on the standby and then applies the config to write config files.
The standby has monitoring disabled until the primary fails and failover is manual... The license agreement says a secondary system shouldn't be actively monitoring or something like that. The cron on the primary would need to be disabled when a failure occurs.

Thanks for the input.

@rajasegar - did you want to keep this open in case anyone else chimes in or are we okay to lock it?

cdienger wrote:@rajasegar - did you want to keep this open in case anyone else chimes in or are we okay to lock it?

Let keep it open for a week.

rajasegar wrote:

cdienger wrote:@rajasegar - did you want to keep this open in case anyone else chimes in or are we okay to lock it?

Let keep it open for a week.

Sounds good