Skip to main content

Troubleshooting

Quick Diagnosis​

# Check container status
docker ps | grep circuit-breakers

# Check recent logs
docker logs circuit-breakers --tail 50 2>&1 | grep -i error

# Check fallback buffer (if using file fallback)
ls -la /var/buffer/failed-requests.jsonl 2>/dev/null || echo "No buffer file"

Common Issues​

Circuit breaker not triggering​

Cause: Threshold not reached or timeout too long

# Check logs for retry attempts
docker logs circuit-breakers --tail 50 2>&1 | grep -i retry

Fix: Lower the failure threshold or reduce timeout:

max_retries: 3
timeout: 5s  # Reduce from 30s

All requests going to fallback​

Cause: Primary destination down or unreachable

# Test primary destination directly
curl -s -o /dev/null -w "%{http_code}" $PRIMARY_API_URL/health

Fix: Verify destination URL, check network, restart destination service

Fallback buffer filling up​

Cause: Primary has been down too long, buffer not being replayed

# Check buffer size
wc -l /var/buffer/failed-requests.jsonl 2>/dev/null

Fix: Fix primary destination, then replay the buffer manually or restart the pipeline

Data loss during failover​

Cause: No fallback configured or fallback also failing

Fix: Add multi-level fallback chain:

output:
  fallback:
    - http_client: {url: "${PRIMARY_URL}"}
    - http_client: {url: "${SECONDARY_URL}"}
    - file: {path: "/var/buffer/emergency.jsonl"}

Still stuck?​

  1. Add debug logging: logger: {level: DEBUG}
  2. Check the Complete Pipeline for reference config
  3. Review Priority Queues for handling degraded service