Discover how we banished intermittent “no such bucket” errors by tuning Envoy’s fail-open vs. fail-close behavior, co-locating rate-limit services to slash cross-LB latency, and ditching K6 for Envoy’s own Nighthawk to stress-test at scale. You’ll pick up real-world trade-offs, debugging hacks, and performance tuning tips for latency-critical systems.
We’ll also dive into using gRPC metadata tagging for better observability—rolled out seamlessly in production. If you’re into resilient, high-traffic architectures, catch this talk at KubeCon + CloudNativeCon NA in Atlanta (Nov 10–13)!
Watch on YouTube
Top comments (0)