Practically speaking, all software systems crash at some point. One of the reasons I love working with serverless systems so much is that they, by their very nature, keep an application relatively small and more akin to a microservice, rather than a broad monolithic application. This fact by itself can drastically reduce the number of ways an application can fail. However, at some point, it will fail, and an exception you didn't expect will occur. What, then, is the best way to handle unexpected exceptions in a serverless system?
The good news here is that we have multiple options, and that some systems you may already be familiar with can work in the same way as they would in a non-serverless system. In the following sections, we'll walk through the steps for integrating two popular error tracking services, Sentry and Rollbar. Both services offer similar functionality and are equally easy to set up. In the following examples, we'll be using Python, but both Sentry and Rollbar...