Help! Something’s Wrong With My Mesh: A Linkerd Troubleshooting Checklist

You've deployed Linkerd, your services are meshed, and mTLS is running. Then something looks off. Traffic errors you can't immediately explain. A pod that isn't injected. A certificate warning you weren't expecting.

Knowing where to look, and in what order, is the difference between a five-minute triage and a two-hour incident. In this session, we'll walk through a practical troubleshooting checklist built from real-world Linkerd production deployments: how to read linkerd check output, how to diagnose proxy injection failures without assumptions about how your team configured it, the three distinct traffic error signatures and how to interpret them, and how to recognize when the destination controller is the real culprit behind FailFast errors at scale.

• A repeatable triage sequence you can use the next time something looks wrong in your Linkerd deployment.
• A clear understanding of the most common failure modes in production BEL clusters.
• Practical diagnostic commands for each failure category, including BEL-specific checks that the core linkerd check won't surface on its own