I have an older version of NPM running where someone set up a useful chart that I'm trying to recreate in our new NPM 10.4 server. I recreated the universal device poller on the new server. In this case it is to monitor the current number of connections per server in a pool on the BigIP F5 load balancer. I matched everything up side by side regarding the poller itself. When I look at the universal device poller summary chart on each server, I get different results. On the new NPM chart, none of the servers have more than 3 connections, on the old chart, which is accurate, they're anywhere from 6-15. The time periods are the same. The polling rate is the same. The chart sample interval is the same. It is not a difference in scale either, meaning one isn't 10x more than the other. I've tried every other type of universal polling chart available to see if I get different results but it is consistently wrong, although I did find a legacy chart that matched the old chart type so I could be sure every setting matched, but the results are wrong.
The older NPM server version is labeled as 2010.1.0.
Anyone have any ideas?
npm