Skip to content

Expose Prometheus/OpenTelemetry metrics for outbox queue depth, processing latency, and dead letter count #22

Description

@Sarmkadan

There is currently no built-in observability into the health of the outbox processor: operators cannot alert on growing queue backlogs, elevated processing latency, or an increasing dead letter count without querying the database directly. The library should emit OpenTelemetry metrics (compatible with Prometheus scraping) including outbox_pending_messages_total, outbox_processing_duration_seconds, outbox_dead_letter_total, and outbox_publish_errors_total. These should be registered automatically when the consuming application has OpenTelemetry configured, following the opt-in pattern used by other ASP.NET Core libraries. This would enable standard Grafana dashboards and alerting without custom instrumentation.

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or request

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions