Logs remain the most widely used telemetry signal, yet managing them at scale is often challenging due to inconsistent formats, bundled payloads, and the need for governance and cost control. The OpenTelemetry Collector provides a vendor-neutral way to ingest, process, and export logs, enabling organizations to build scalable and manageable pipelines.

This session will explore how to design log ingestion pipelines utilizing the Collector’s built-in processors for filtering, enrichment, and transformation, ensuring logs are manageable and scalable across diverse environments. We will highlight the new Unroll Processor, which expands bundled log records (such as JSON arrays) into individual events, making downstream analysis more reliable. Attendees will also learn how to normalize structured, semi-structured, and unstructured logs, apply transformations with the OpenTelemetry Transform Language (OTTL), and enrich telemetry with metadata for better context.

Attendees will learn how to design log pipelines using the OpenTelemetry Collector processors. They will leave with practical strategies to normalize, transform, and manage logs efficiently across distributed systems.