The Pod Was Cheaper. The Service Wasn’t.
Use OpenCost and proxymock to prove Kubernetes rightsizing lowers cost per successful request without hiding behavior or throughput regressions in testing.
Posts about observability
Browse 28 posts in this category
Flamegraphs Find It. Replay Proves It.
Use Grafana Pyroscope and proxymock with an AI coding agent to find a Go CPU hotspot, preserve API behavior, and verify the performance fix.
eBPF: Preventing Garbage HTTP Payloads When Reading Kernel Scatter-Gather Buffers
How our nettap eBPF agent produced garbage HTTP bodies from kernel iov_iter scatter-gather buffers, and the CO-RE plus task-local storage fix that made it correct.
AI writes code in seconds, but delivery still takes days
AI pushed throughput up 59%, yet median delivery got worse. The bottleneck moved to validation. How replaying production traffic in CI closes the gap.
Trace without traces
The trace was sampled out. I found the bug anyway — by filtering recorded traffic on the customer's email instead of a trace ID. Here's how to follow one request across four services with no trace IDs and no OpenTelemetry.
The Three Pillars Were Built for Humans
Metrics, logs, and traces were built for humans and cheap storage. AI inverts both assumptions, and the next maturity level is a deterministic replay sandbox.
Five things your logs will never tell you
Logs, metrics, and traces are a lossy compression of production. Five things you can do with a traffic data lake that observability can't.
Logs told me something broke. Traffic showed me what.
Capture production traffic and store it in your own Elasticsearch with Speedscale BYOC. Pull it locally with es-gather.py and reproduce bugs with proxymock.
The Kubeshark Workflow That Doesn't Stop at the Dashboard
A Kubeshark alternative that goes beyond observability. Stream live cluster traffic into proxymock, then replay or mock it locally from your laptop.