With this approach, we expose one or more endpoints of a service to emit diagnostic information related to the service itself as well as the infrastructure parameters. Generally, different endpoints focus on providing different information. For example, one endpoint can give the health status of the service, while the other could provide the HTTP 500 error information that occurred in that service execution. This is a very helpful technique for microservices since it inherently changes the monitoring from being a push model to a pull model and reduces the overhead of service monitoring. We can scrap data from these endpoints at a certain time interval and build a dashboard and collect data for operational metrics.