Collect prometheus snapshots #83
Description
In a cluster that has prometheus running in it, we should collect the snapshots of the databases for analysis offline. To analyse locally, we can start prometheus as a docker container for example and mount a volume with the snapshot to have prometheus expose use that data. From there we can run various queries.
Considerations
- When creating a snapshot, use
skip_head=<bool>to ignore data that has not been persisted to disk by prometheus - Have a short writeup on how to use the database dump
- Write up on a few queries such as graphs of disk/cpu/memory utilisation, etcd metrics etc