Hi there,
If we have connectionPool settings in our DR, do we need to do anything special for canary releases to work? If I change the traffic split from 90-10 to 80-20, for example, and at the same time, we’re increasing the replicas value in v2 deployment and decreasing it in v1 deployment to match up with the respective traffic loads, would the connection pool just work seamlessly? Or would we need to worry about forcefully purging it (somehow) to get rid of the now-stale connections (from the v1 pods that got killed) and make way for the new connections (new v2 pods)?
Or should we simply set the connection pool size to the sum of the v1 and v2 deployment replicas so that we don’t have to worry about not having enough space for new v2 connections?
Thanks for your thoughts and general advice on this.