Devops is currenly my bread and butter. For the different infra I manage I like to collect metrics using Prometheus.
When it came time to design and host a Tezos baking infra I wanted to collect baking metrics using Prometheus. As I could not find any solution already, I made one 😬🙌🌈🚀
I collect metrics from the RPC API of my own Tezos node(s). It's quite easy to work with and figure out 👍
It's still a WIP though, I know a couple of issues with how it collected metrics and presents them - but it work well enough to make informed decisions about the state of your node and baker 👍