TCP/UDP Monitoring
Monitor services at the network layer
Monitoring
TCP and UDP monitors check connectivity at the network layer by attempting to connect to a specific port. These are ideal for monitoring non-HTTP services like databases, mail servers, game servers, and custom applications.
TCP vs UDP
TCPConnection-Oriented
TCP monitors establish a full connection handshake to verify the service is accepting connections.
- ✓ Reliable connection verification
- ✓ Detects service crashes
- ✓ Works with most services
UDPConnectionless
UDP monitors send packets and check for responses. Best for services that use UDP protocol.
- ✓ DNS servers
- ✓ Game servers
- ✓ VoIP services
Configuration
Host
The hostname or IP address to monitor.
db.example.comPort
The port number to connect to.
3306 (MySQL)5432 (PostgreSQL)6379 (Redis)27017 (MongoDB)
Timeout
Maximum time to wait for connection (in seconds). Default: 10 seconds.
Common Use Cases
| Service | Default Port | Protocol |
|---|---|---|
| MySQL | 3306 | TCP |
| PostgreSQL | 5432 | TCP |
| Redis | 6379 | TCP |
| MongoDB | 27017 | TCP |
| SSH | 22 | TCP |
| DNS | 53 | UDP/TCP |
| SMTP | 25, 587 | TCP |
Best Practices
- Use TCP monitors for connection-oriented services (databases, mail)
- Ensure firewall rules allow monitoring from your ATStatus instance
- Set appropriate timeouts based on network latency
- Consider using database-specific monitors for deeper health checks