API Monitoring - How deep do you go?

Google Cloud Apigee
1

When setting up monitoring for your API Program what has been the approach to monitoring depth? By that I mean have you just set up monitoring for the proxy itself or does your monitoring of the API also cover the health of the underlying target end point?

More ideas here but I would be interested in the approaches taken by the community.

2

We are doing pings of resources and status of the cloud. we are also implementing health checks and log monitoring. We are struggling with determining how much to monitor and logs to collect. I’m pushing for a just enough monitoring and data but drawing that line is difficult.