open-telemetry/opentelemetry-collector-contrib

The OpenTelemetry Collector Contrib repository serves as a centralized hub for OpenTelemetry Collector components that fall outside the scope of the core repository. Written in Go, this project hosts a diverse collection of receivers, processors, and exporters that enable organizations to collect, process, and export telemetry data including traces, metrics, and logs across distributed systems. The repository explicitly distinguishes itself from the core collector by housing components deemed unsuitable for inclusion there, though some widely-used components like Jaeger and Prometheus exporters are distributed as part of both core and contrib releases through the opentelemetry-collector-releases repository.

The project emphasizes flexibility in deployment through the OpenTelemetry Collector Builder, which allows users to construct custom distributions by selecting components from the core repository, contrib repository, or third-party sources. Each component maintains its own stability level per signal type, meaning a single component might be stable for traces while remaining in alpha for metrics or development stages for logs. This granular stability model sets clear expectations for users depending on specific components for production workloads.

The most active labels tracked are enhancement with 1807 items, Stale with 1758 items, and bug with 1577 items, indicating ongoing feature development alongside maintenance of existing functionality.

The repository maintains a structured governance model with distinct roles including maintainers, approvers, and triagers drawn from organizations like Splunk, Elastic, Honeycomb, Dynatrace, DataDog, and Grafana Labs. Support for individual components is distributed across community maintainers and vendor-specific teams, with the broader maintainer group retaining authority to downgrade components deemed unmaintained or posing risks to the distribution. The project explicitly notes that actual support typically flows from individual contributors and code owners rather than centralized teams.

Feature gating mechanisms allow experimental functionality to be hidden behind gates before integration into main code paths, with these gates themselves subject to defined lifecycle stages. The repository's classification spans processors, telemetry collection, tracing, logging, instrumentation, exporters, observability, receivers, and metrics, reflecting its comprehensive role in the observability ecosystem.