On 24 May 2010, at 15:45, Andrew Hall wrote:
Thought this might be of interest to the community... http://www.shinken-monitoring.org/features/ I believe it was featured in Linux Magazine recently. Are any of the extra features of benefit ?
Aside from being based on Python, which is an implementation choice, the main feature is sending out to satellite pollers to do the active monitoring for the master to take the results.
Our solution for that is to use the master/slave architecture to do that. The "poller" is effectively another Nagios instance and I don't see any benefit to their way versus ours. I accept there are some limitations with our distributed architecture choices (namely the amount of output sent back from a slave), but we would be swapping "known but overcomeable limitations" with "unknown limitations" with a change of this magnitude.
I think the issue is a single, actively polling Nagios instance doesn't scale for lots of hosts. But we think a single Nagios master (the Opsview master) that processes just passive results does scale up to the tens of thousands of hosts, so that's where we're betting the future will be.
Ton
_______________________________________________ Opsview-users mailing list Opsview-users@lists.opsview.org http://lists.opsview.org/lists/listinfo/opsview-users
