Collect operator autopilot health metrics

[chuckyz]

This extends the exporter to collect everything exposed by the https://www.consul.io/api-docs/operator/autopilot#read-health endpoint. Naming is weird, but I wanted to call out that this isn't something that should typically be collected unless you need to keep track of really low level stats.

This is my first time extending a prometheus exporter and I find this one super helpful, thank you all for all your hard work and please let me know if anything here is super wrong. :)

[chuckyz]

Ignore previous iterations of this comment I was building off my local master... 🤦

Fixing test issues and lint problems now.

[simonpasquier]

Thanks for your contribution. A few questions/remarks but nothing blocking.

[chuckyz]

I had to soft rebase off origin's head, and then git commit --signoff followed by git push +operator_autopilot_health.

For anyone who journeys here and hits something similar, hopefully https://gist.github.com/patik/b8a9dc5cd356f9f6f980#the-harder-and-less-flexible-way helps to clean up.

Also, for DCO bot whatever reason commits from my Windows desktop with my main email never pass the DCO check.

[simonpasquier]

Apologies for the very belated answer. Would you be able to rebase?

[chuckyz]

@simonpasquier I can.

In Consul 1.9 they also added a lot of new autopilot metrics that are extremely useful. Do you have a way to deal with version breakage?