Coordinator and Overlord Readiness Checks

[Julian Jaffe]

Hey all,

I'm looking to add readiness checks to coordinator and overlord nodes to
improve automatic deployment of clusters. More specifically, I'm planning
to add an endpoint to the coordinator and overlord that returns 200 OK when
the node is ready to process lookup/tiering rules/supervisorspec/etc.
requests. My first thought is to add another module and bind at the
Stage.ANNOUNCEMENT stage. Is there any downside to this approach, or were
readiness checks for historicals implemented differently for historical
reasons?

Thanks,
Julian