Patterns for resilience
When we are thinking about the resiliency of the application, we should try to answer the following questions: Can the application handle failure conditions? If one component of the application fails, does it bring down the whole application? Is there a single point of failure in the application?
Let's look at some patterns that will help us to make our application resilient.
The circuit-breaker pattern
This is an important pattern to implement both resilience and responsiveness in the system. Often, when a service fails in a system, it impacts other services as well. For example, service X calls service Y in the system to get or update some data. If service Y is unresponsive for some reason, our service X will make a call to service Y, wait for it to timeout, and then fail itself. Think of a scenario where service X itself is called up by another service P, and so on. We are looking at a cascading failure here, which will eventually bring down the whole system.
The circuit...
