Cacti instances, one minute and five minute

[logisist]

I have a box with eight cores, 16GB of RAM, and 900GB of RAID5, 15k space. So, server is goood. eh? At least for the ~2000 devices I'm looking to monitor. Distibution network devices (<60) at one minute and (>1900) at five minutes.

My question is about crontab jobs for the two Cacti instances I intend to run; one with a five minute poll and the other with a one minute poll... since they're running in parallel on the same box, with different spine configs, what should I do for my crontab entries?

I've created two databases, cacti-0.8.8a-1m and cacti-0.8.8a-5m along wtih /etc/spine-0.8.8a-1m.conf and /etc/spine-0.8.8a-5m.conf.

There are two databases; cacti-0.8.8a-1m and cacti-0.8.8a-5m.

I haven't yet ventured into the arena of the Cacti data-source at 1 minute, etc...

[phalek]

I'm not quite sure why you want to have two cacti instances running on a single server. Actually, you can have a mixed Cacti system running 5m and 1m polls at the same time. Cacti will make sure the graphs/devices are polled with their correct polling interval.

Cacti EZ is setup if you want to have a reference system.