Nagios isn't the best, it's fragile, and Nagios 3 won't do anything to change it.
There are other packages (Zabbix and Zenoss look good) out there, so we should look at these instead of rewriting the nagios configs for nagios3 and finding out it's still as ugly :).
I'm acquired publictest3.fp.o so we can test these packages, but we should work out exactly what our requirements are, as such we'll use the last half hour of the Infrastructure Meeting to discuss.
I've used Zabbix quite extensively at $DAYJOB and like it quite a bit. It's already in Fedora too. What I like the best is that it combines graphing and monitoring/notification all into one package. Unfortunately I won't be able to be at the meeting this Thursday...
I also support migrating to Zabbix. I've also used it at $dayjob and think it offers some excellent features. I'd be happy to help set it up and get it configured.
Short discussion was held, summary/logs are at: https://www.redhat.com/archives/fedora-infrastructure-list/2008-July/msg00143.html
We've not deployed zabbix completely but we have picked it and it is installed and running. Closing this ticket.
to comment on this ticket.