Docker HEALTHCHECK guidance

[codefromthecrypt]

There is awesome flexibility we allow with anything from all-in-one until distributed deployment. I love it.

This presents a healthcheck challenge as certain ordering constraints remain and we want staged things to work naturally. Consider the case of ServiceManager starting controller, server and broker. We'd want one "HEALTHCHECK" to check regardless of the process config.. ideally an HTTP path like /health

can we do this?

[kishoreg]

providing a /health end point should be possible, we also need a swagger resource

[codefromthecrypt]

I think for servicemanager, we should have a composite health returned. there's a conversation on slack to this effect