01
Platform
Orchestration, probes, and the failure of a control plane.
3Every platform incident p99 has published — what broke, what the operator was told, and the fix that held.
02
Incidents
3- No deploy, no traffic change — every internal call fails at exactly 00:00:07tls · mtls · certificates · expiry · observabilityeasy
- One customer's 50,000 req/min spreads across 20 instances. Only shared state sees the whole stream.redis · rate-limiting · distributed-systems · concurrency · slahard
- Traffic spikes. Pods start dying. Healthy pods, killed by their own cluster.kubernetes · cascading-failure · load-sheddinghard
→