life-my--midst--in

Infrastructure

Infrastructure-as-code and deployment configuration:

• terraform/ – Cloud resources
• helm/ – Helm chart scaffold (values.yaml holds env overrides)
• k6/ – Load testing scenarios (see k6/README.md)
• grafana/ – Dashboards and provisioning configuration
• prometheus/ – Prometheus scrape configuration
• docker-compose.yml – Local dev stack (Postgres, Redis, API, Orchestrator, Web)
• docker-compose.prod.yml – Production Compose configuration
• Dockerfile – Root web application image (multi-stage Node 22 Alpine)
• railway.json / render.yaml / vercel.json – PaaS deployment configs
• .env.example / .env.integration.example / .env.production.example – Environment templates

Migrations & Seeds

• Local: docker-compose -f infra/docker-compose.yml run --rm migrations (profile init) runs API + Orchestrator migrations/seeds against compose Postgres.
• CI: .github/workflows/test.yml runs migrations before tests; integration jobs bring their own Postgres/Redis.
• Helm: values.yaml includes migrations.env; templates/migrations-job.yaml installs as a pre-install,pre-upgrade hook and runs pnpm --filter @in-midst-my-life/api migrate && pnpm --filter @in-midst-my-life/orchestrator migrate.

Environments

• API: DATABASE_URL/POSTGRES_URL required when using Postgres repos.
• Orchestrator: DATABASE_URL + ORCH_TASK_STORE=postgres, TASK_QUEUE=redis, REDIS_URL.
• Web: NEXT_PUBLIC_API_BASE_URL, NEXT_PUBLIC_ORCH_BASE_URL.

Keep dev/test/integration databases distinct (e.g., midst_dev, midst_test, midst_integration). Seeds are idempotent (ON CONFLICT DO NOTHING).

Load Testing

Performance testing is done with k6. See k6/README.md for detailed instructions.

Quick start:

# Install k6 (macOS)
brew install k6

# Run smoke test (quick validation)
k6 run infra/k6/scenarios/smoke.js

# Run baseline test (normal load)
k6 run infra/k6/scenarios/baseline.js

# Run stress test (find breaking point)
k6 run --env SCENARIO=stress infra/k6/scenarios/baseline.js

Monitoring

Grafana dashboards are in grafana/dashboards/:

• api-metrics.json – Comprehensive API metrics (latency, throughput, errors)
• api-performance.json – Performance-focused view
• database-redis.json – Database and cache metrics
• system-overview.json – System-level metrics

Import dashboards via Grafana UI or use the provisioning files in grafana/provisioning/.

This site is open source. Improve this page.