This message was deleted.

We handle stuff like this in our CICD process via bash scripts. I’ve never worked in Azure so not sure what the best path might be but you could potentially

curl

the WebApp and set a conditional based on that or maybe use the Azure CLI to determine what to do next?

Broadly (!) it makes sense to run a smoke test of some sort after any deployment even if that check is only to ping a healthcheck endpoint or equivalent. Addressing that takes you a good step towards the "is it running" problem

I fully agree!! But depending on the outcome of this test, I might wanna react in my pulumi program; e.g. in a blue/green deployment, I don’t want to switch to the new instance if it is not up and running. But my feeling is, that pulumi is not very helpful to this kind of thing - although I would love to be proven wrong!

Ah... there are all kinds of interesting questions that arise from that - about what should be orchestrating which, when, and how. The long term answer is that the Automation API - running pulumi from code rather than the command line - should let you address that. In the short term you're ahead of me in terms of deployments anyway so I've hit the limits of practical experience...