Operations

This section covers day-to-day operational procedures for running SLOzy in production.

Contents

• Monitoring — Health checks, Prometheus metrics, cache statistics, real-time SLO metrics
• Performance — Caching, connection pooling, query optimization, rate limiting, build optimization
• Backup & Recovery — PostgreSQL backup, migration management, Kubernetes CronJob automation

Key Operational Requirements

Requirement	Minimum	Recommended
PostgreSQL	16+	16+ with replication
Redis	7+	7+ with persistence
Disk	1 GB	5 GB +
Memory	512 MB	1 GB
CPU	1 core	2+ cores

Health Check Endpoints

SLOzy exposes two health check endpoints used by load balancers and orchestrators:

• GET /health — Liveness probe (basic process health)
• GET /api/v1/health — Detailed readiness probe with service dependency checks

Both return JSON with status, timestamp, and (for the detailed endpoint) per-service health status.

Monitoring Stack

The recommended monitoring stack includes:

• Prometheus — Metrics collection and alerting (runs on port 9090)
• Grafana — Dashboards and visualization (runs on port 3001 by default)
• Alertmanager — Alert routing and notification (port 9093)

See Monitoring for detailed configuration.