Performance Monitoring for ManageIQ


#1

Hi Everyone,

I have put together an easily deploy-able performance monitoring solution for ManageIQ. The solution uses Ansible to deploy Collectd, Graphite and Grafana on your appliance and saves metrics at a 10s interval into a TSDB (Time Series Database). Not all of the kinks have been worked out, but it would be great to get some community feedback on this type of solution.

Also included are several performance monitoring dashboards tailored toward the provider type you are managing and a comprehensive dashboard showing everything as well.

You can find the instructions to deploy it on your ManageIQ appliance here.

Attach are two screenshots giving you a taste of all the data you will get insight into.

CPU/Memory Example:

Per Process CPU usage example:

Hope you all found the insight to this data as useful as I have.

-Alex


#2

Once we switch the backend for metrics to a new TSDB, I think it would be cool if these metrics just lived in that database.

Then, we could bake in the diagnostics UI directly, much like we had started to do in the “database diagnostics” section.

Really cool @akrzos !