Observability Specification for IronCore components

Summary

"Observabilty" consists of metrics and logs that provide insight into a component's current state and possibly interplay with other components.

We need to define a common way to expose, collect and process such information.

Scope

In Scope

• Decide on a collection direction for metrics (push vs. pull)
• Define a transport mechanism for metrics
• Define guidance, a structure and approach for logs (log levels, log format/structure, where to write logs that they can be collected centrally)

Out of Scope

• Software stack for collecting, processing, visualizing metrics or logs
• Deployment considerations for an observability stack (location, backup, retention, etc)

Acceptance Criteria

• The transport format for metrics is defined
• The collection mechanism for metrics (push vs. pull) is documented with best practices
• The recommended format for logging is defined
• Guidance on logging libraries to be used is provided
• General best practices on naming (metrics, logging fields) for consistency is provided