Wire in isAlive middleware to loop-check pod status

[UcDust]

So this patch has been deployed. You can see a partial example here: https://pgfarm.library.ucdavis.edu/api/db/library/foo?isAlive=true where the pod status is returned. If a db is not responsive on a POST/PUT a 503 is returned and the payload will contain the podStatus will all sorts of information about the pod.

Can one of you implement a middleware on the client side that retries URL requests when at 503 is returned? Perhaps up to 10 times? 3 second delay? That would give a 30 second repair window. If after the 10 attempts, you can then pass along the last request as the final state. Additionally, we could provide a secondary response channel (event emitter or otherwise) to send along pod status updates. We will have some k8s details we could theoretically display on screen, let the user know things are happening. But that might be a feature for another day