Bitbucket Pipelines now emits OpenTelemetry traces via webhook events, enabling teams to stream detailed pipeline execution data into their observability stack. The trace hierarchy covers four span types: pipeline run (root), step, command, and container resource metrics. Each span carries structured attributes for timing, state, trigger type, and CPU/memory usage. Setup involves configuring a webhook in repository settings to point to an OTel-compatible ingestion endpoint. Once wired up, teams can identify slow commands, detect flaky steps, profile container resource usage, and correlate pipeline runs with application traces using shared identifiers like build number or branch.

6m read timeFrom atlassian.com
Post cover image
Table of contents
Why pipeline traces?How pipeline traces are deliveredThe span modelConsuming traces in your observability stackPutting it together: typical use casesWhat better visibility means in practice

Sort: