Allow VIP swap when either of deployments is recycling
Suppose I have a bug that suddenly manifests itself in production deployment and this causes some instances to fail and recycle. I identify the bug, fix it in my code, prepare a new service package and deploy it into staging. I then try "swap VIP" but it fails all the time - that's how it currently works. Which means I cannot replace my faulty code with good code without downtime. VIP swap primary purpose is to prevent downtime and this is the scenario where it is needed the most but it doesn't work here.