Monitoring

Good monitoring is important to keeping any application healthy and available. In a microservices architecture good monitoring is even more important. We not only need to be able to monitor what’s going on inside a service, but also the interactions between the services and the operations that span them.

• Use activity or correlation IDs: These are used to associate events to a specific activity or transaction.

• Use a common log format: This enables better correlation of events across services and instances.

• Collect and analyze logs: By collecting logs from the various services and instances, we can analyze and query across multiple services and instances of a service.

• Consider using a sidecar: A sidecar is simply an out-of-process agent that is responsible for collecting metrics and logs. We can run this in a linked container.

In addition to these points, all the standard monitoring tools and techniques should be used where appropriate. This includes endpoint monitoring and synthetic user monitoring.

See Chapter 7 for more information on best practices for monitoring and logging with microservices architectures.