Distributed monitoring with centralized config management
Posted: Wed Aug 29, 2012 2:47 pm
I currently run a Nagios 3.x setup where I have 3 servers that handle monitoring for 3 different locations with 2 setup as slaves to the 3rd, and maintain a single central configuration repository using a hacked version of Groundwork Monitor Architect. I'm interesting in upgrading to Nagios XI (or possibly Core), but would like to maintain the same kind of configuration as I currently have. Is that possible?
For reference, here's how I have things set up:
Slave servers:
Monitor the local network
Use OCHP and OCSP to send check results via NSCA
Perform active checks
Notifications are disabled
Master server:
Monitors local and remote networks.
Performs active checks for local hosts/services and passive checks for remote
Receives remote check results via NSCA
Maintains central configuration and pushes subset configurations to remote systems
Adjusts parent/child relationships when pushing remote configurations
Sends all notifications
I am not interested in using DNX to randomly distribute checks among servers, as the whole point is to have active checks run only on the local network unless there is a problem with the local monitoring system. I know I can fairly easily set up a central dashboard for results, what I don't know is if/how to set up central configuration.
TIA for the suggestions!
For reference, here's how I have things set up:
Slave servers:
Monitor the local network
Use OCHP and OCSP to send check results via NSCA
Perform active checks
Notifications are disabled
Master server:
Monitors local and remote networks.
Performs active checks for local hosts/services and passive checks for remote
Receives remote check results via NSCA
Maintains central configuration and pushes subset configurations to remote systems
Adjusts parent/child relationships when pushing remote configurations
Sends all notifications
I am not interested in using DNX to randomly distribute checks among servers, as the whole point is to have active checks run only on the local network unless there is a problem with the local monitoring system. I know I can fairly easily set up a central dashboard for results, what I don't know is if/how to set up central configuration.
TIA for the suggestions!