Why is Endpoint Monitoring so Important?
The concept of EndPoints is important when monitoring microservices-based applications (actually, SOA applications, too). A good way to think about it is that a service is really just an aggregation of endpoints. You read that right, a service can be thought of as an aggregation of all the endpoints delivering the functionality of that service:
- URL's used for web service calls
- queries to a database service
- many other kinds of calls
When trying to understand the performance and health of a service, it’s the endpoints that really matter. Which endpoint is slowing down? Is there an endpoint throwing an excessive number of errors? Which endpoint is using the most resources?
Automatic Endpoint Monitoring
In our latest release, Instana automatically discovers and monitors service endpoints. No user setup is required. Endpoints are automatically labeled for monitoring based upon various rules.
This new level of automation greatly enhances the understanding provided by Instana while holding true to the mantra of “Maximum Understanding, Minimum Effort”.
Using Service Endpoints
The total number of endpoints per service is shown in Instana's Service View Dashboard. The image below shows all services sorted by the number of endpoints per service.
To see the data for an endpoint, you’ll need to access a Service View either from within or outside of an Application Perspective. Click “Endpoints” to access the Endpoints View, which is a tabular list of all endpoints. In a modern microservice application, this could be a short list with only a few entries. In an older SOA environment, this list could be much longer with a single service supporting many different endpoints.
This table makes it easy to quickly identify the endpoints with the most calls, highest latency, or most errors.
After you have identified an interesting endpoint, click on the endpoint name to shift focus to that individual endpoint and arrive at the summary screen. All dashboards are now updated to show information relevant to that endpoint.
From here, you can troubleshoot the root cause of any slowness or errors.
The automation of service endpoints is a significant capability that will save you time and will improve your understanding of the monitored environment.