Building Microservices with Go

Monitoring/alerting

When the application is running, we need to be sure of the health and status of the application. Monitoring is an incredibly important facet of the continuous deployment life cycle. If we are deploying automatically, we need to understand how our application is performing and how this differs from the previous release. We have seen how we can use StatsD to emit data about our service to a backend such as Prometheus, or a managed application such as Datadog. Should our recent deploy exhibit anomalous behavior, we are alerted, and from there we can act to help identify the source of the problem, intermittently rolling back if necessary or modifying our alerts because our server may just be doing more work:

# Create a new ...