ryan.gogo eb6b6e09df Add health checks for zero-downtime deployments ...

Critical changes for production stability:

1. Add healthchecks to all services:
   - app: curl to /health endpoint (10s interval, 5s start period)
   - postgres: pg_isready check (5s interval)
   - redis: redis-cli ping check (5s interval)

2. Use depends_on with health conditions:
   - Ensures postgres/redis are fully ready before app starts
   - Prevents connection errors during startup

3. Remove hardcoded container_name:
   - Prevents name collisions between deployments
   - Allows Coolify to manage container names

Why this matters:
- Without healthchecks: Coolify kills old container → builds new → 502 gap
- With healthchecks: Coolify builds new → waits for healthy → kills old → zero downtime

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

2026-02-15 10:54:44 +01:00

1.9 KiB

Raw Blame History

View Raw