There are some expectations around the monitoring, that is to understand concepts such as liveness and readiness
- Liveness - Is the current running state and health of the application service
- Readiness - This is when the service is ready to be consumed. An example is performing some schema migrations or waiting for dependent services to be ready before your service can be ready. We wouldn't want the service to be consumed until it is ready to be so.
- A
/metrics
endpoint for your service to expose all necessary metrics - A
/healthz
endpoint for the liveness Probe - A
/readiness
endpoint for the readiness probe