Log Support in OpenTelemetry

Open Telemetry is an open source and vendor agnostic technology that standardizes telemetry data generation, collection, and exportation to various backends. It consists of three major components: specification, instrumentation libraries, and collector. The log data model is flexible and supports many fields that can be defined, but none of them are required. Context is a key concept in Open Telemetry, allowing for the stitching of all telemetry data together. Resources and attributes are also foundational pieces of Open Telemetry, providing a normalized way of understanding unique objects in an environment.

• Open Telemetry standardizes telemetry data generation, collection, and exportation to various backends
• Consists of three major components: specification, instrumentation libraries, and collector
• Log data model is flexible and supports many fields that can be defined, but none of them are required
• Context is a key concept in Open Telemetry, allowing for the stitching of all telemetry data together
• Resources and attributes provide a normalized way of understanding unique objects in an environment

In a distributed tracing environment, Open Telemetry allows for the association of logs with their associated traces, providing core context between signals.

OpenTelemetry’s charter is to provide an observability framework for cloud-native software. Its goal is to provide full support for traces, metrics, and logs and provide a single implementation that can be leveraged. Traces have already reached a stable maturity level and metrics are well on their way. While logs were outside the GA scope of the project, they are an important data source and initial work has already kicked off to support them. - An initial data model has been accepted by the project - An initial implementation exists in the Java client library - Initial support has been added to the collector End users are interested in the plans for logs in OpenTelemetry. Will logging client libraries be introduced? Will native logging capabilities be added to the collector? How do context and resources work with logs? This session will dive into these topics to explain the current state and future direction. A demo of the current capabilities will be provided.