Root-Causing Incidents Without Re-Deploying Production - Using bpftrace for Kubernetes Debugging

Explore advanced techniques for troubleshooting production incidents without redeploying in this 30-minute conference talk from KubeCon + CloudNativeCon Europe. Learn how to leverage bpftrace to capture crucial data such as function arguments, return values, and latencies when logs fall short. Discover powerful strategies for root-causing issues in Kubernetes environments, including tips for overcoming the complexities of multi-node setups. Follow along with a demo incident to gain practical insights on deploying and collecting data from bpftrace scripts using Pixie. Enhance your ability to diagnose and resolve production problems efficiently, minimizing downtime and reducing the risks associated with redeployments.

When the Logs Just Don’t Cut It: Root-Causing Incidents Without Re-Deploying Prod- Phillip Kuznetsov