Operations
Production deployment and operations guides.
Reliability-first operating model
Section titled “Reliability-first operating model”EdgeQuake now documents and follows a few simple operational invariants:
- pin the Rust toolchain so local results and CI results do not drift
- use readiness probes instead of fixed sleeps when starting services
- cancel superseded CI runs on the same branch to reduce stale signal and wasted minutes
- keep heavyweight coverage and full-E2E flows outside the fastest blocking feedback loop
- fail closed when an explicit workspace context is invalid or missing
Guides
Section titled “Guides”- Deployment — Docker, Kubernetes, and bare-metal deployment.
- Configuration — Environment variables and runtime settings.
- Monitoring — Health checks, metrics, and observability.
- Performance Tuning — Optimize throughput and latency.
- Metadata Debugging — Inspect and debug extracted metadata.