AKAyush Kaushikinlogiclooptech.dev·Feb 24 · 5 min readFastAPI SQLAlchemy Session Leak Detection: Diagnose and Fix Long-Running DB Sessions in ProductionYou deploy your FastAPI application. Memory usage is stable. Latency is low. Two days later, your database CPU spikes to 100%. Your logs scream TimeoutError: QueuePool limit of size 5 overflow 10 reac00
AKAyush Kaushikinlogiclooptech.dev·Feb 16 · 4 min readFastAPI Session Leak Detection: How to Find and Fix Long-Running SQLAlchemy Sessions in Production?You deploy your FastAPI app.Everything looks healthy. Latency is low.Memory usage is stable.Health checks are green. Then two days later: • Database CPU spikes• Requests start timing out• Logs show QueuePool limit reached• Restarting pods “fixes” it…...00
AKAyush Kaushikinlogiclooptech.dev·Feb 14 · 4 min readWhy Your FastAPI App Works Locally But Fails in ProductionThe most dangerous phrase in software engineering is: “But it works on my machine.” On your laptop, everything feels perfect. Your FastAPI service runs on a MacBook with plenty of RAM, near-zero latency, and one user: you.Tests pass. Endpoints resp...00
AKAyush Kaushikinlogiclooptech.dev·Feb 11 · 4 min readWhy Uvicorn Health Checks Fail Under Load (And How to Fix It Properly)You deploy your FastAPI app to Kubernetes (or ECS).You configure a liveness probe on /health.Everything looks green. Then traffic spikes. Pods start restarting. CrashLoopBackOff. You check logs: No Python exceptions No OOM CPU high, but not maxed ...00
AKAyush Kaushikinlogiclooptech.dev·Feb 6 · 6 min readFastAPI in Production: The Complete Deployment Guide (Docker, Workers, Scaling & Best Practices)You built a blazing-fast FastAPI service locally.Tests pass. Swagger looks clean. Latency is beautiful. Then you deploy it. And suddenly the real problems begin. Workers crash without warning.Connections exhaust.Latency spikes even though CPU looks i...00
