Monitoring and Observe-ability

[a-musing-moose]

🎯 Aim

Ensure that the projects that are created from this template are tooled up to be easy to monitor and observe

📕Context

In order to be able to effectively maintain software, diagnose and resolve issues and preempt performance issues built in support for observe-ability is key. There are lots of different approaches but at least in the Python world Sentry is a tried and trusted approach to "crash-lytics. Beyond that the space is a little more open, but we should aim to support the adding a greater degree of monitoring than just exception tracking. Ideally up to and including custom events and metrics

📝 Relevant resources/doc's/people

• Sentry - Exception tracking amongst other things
• OpenTelemetry - Open source approach to observe-ability and monitoring. Supported by a number of vendors including Honeycomb, Datadog, Sentry, and New Relic to different degrees.

✅Acceptance Criteria

• All uncaught exception are tracked
• Application performance monitor is possible
• Application tracing is supported
• Custom event and metrics can be added on a project-by-project basis
• Nice to have: Vendor agnostic monitoring/observe-ability to allow us to tailor to customer needs

[heykarimoff]

@a-musing-moose I would like to pick up this ticket. I know about Sentry but less about OpenTelemetry. First, I start reading about it and see what's needed.

[a-musing-moose]

@heykarimoff Awesome - no need to do everything at once. So yeah, go ahead and do a sentry PR first. Then follow up with taking a look at open telemetry as well as perhaps structured logging.