I would consider the term 'API monitoring' as to monitor all the response times and errors on a specific API of all incoming requests on that API.
There are two ways to achieve visibility by applying monitoring to an API:
- Preferred one: Install a Dynatrace OneAgent on the hosts and instances that serve the API in therms of services and therefore get full visibility into all incoming requests on that API in real time. You get all the response times, errors and backend pathes for all API requests that are coming in.
- Synthetic monitoring of an API, which means to trigger the API as a synthetic client and measure and validate the response of your API. Dynatrace offers WebChecks for synthetic monitoring complete Web applications. There is so far no API synthetic test feature within Dynatrace but that is on the roadmap and will be added to our synthetic offering.
Why is the first the preferred one: Because only by monitoring your customers API requests you get a full picture of how your API is running in real scenarios. A synthetic tests always only checks what you defined and will never fully cover the real world API call variations in terms of parameters that your customers are sending in.
