TCP Monitor

Monitor TCP port availability for databases, SSH servers, custom services, and any protocol that runs over TCP. Verify connectivity and measure response time.

What is a TCP Monitor?

A TCP Monitor attempts to establish a TCP connection to a specified host and port. If the connection succeeds, the check passes. If it fails or times out, the check fails.

Use Cases

• Database Servers: PostgreSQL (5432), MySQL (3306), MongoDB (27017), Redis (6379)
• Remote Access: SSH (22), RDP (3389), VNC (5900)
• Mail Servers: SMTP (25, 587), IMAP (143, 993), POP3 (110, 995)
• Custom Services: Microservices, game servers, message brokers
• Infrastructure: Load balancers, proxies, firewalls

Configuration Options

Basic Configuration

Common Ports

Databases

Remote Access

Mail Servers

Message Brokers & Queues

Configuration Examples

Monitor PostgreSQL Database

{
  "type": "tcp",
  "name": "Production PostgreSQL",
  "host": "db.example.com",
  "port": 5432,
  "interval": 60000,
  "timeout": 10000,
  "locations": ["us-east", "eu-west"]
}

Monitor SSH Server

{
  "type": "tcp",
  "name": "SSH Server",
  "host": "server.example.com",
  "port": 22,
  "interval": 300000,
  "timeout": 10000
}

Monitor Redis Cache

{
  "type": "tcp",
  "name": "Redis Cache",
  "host": "redis.example.com",
  "port": 6379,
  "interval": 60000,
  "timeout": 5000
}

Monitor Custom Microservice

{
  "type": "tcp",
  "name": "Payment Service",
  "host": "payment-service.internal",
  "port": 8080,
  "interval": 30000,
  "timeout": 10000
}

Best Practices

1. Choose Appropriate Timeouts

Set timeout based on expected latency:

• Local network: 5 seconds
• Same region: 10 seconds
• Cross-region: 15-30 seconds

2. Combine with Application-Level Checks

TCP monitors only verify connectivity. For databases, also use HTTP monitors to check application-level health:

• TCP Monitor: Port 5432 is open
• HTTP Monitor: /health/db endpoint verifies query execution

3. Monitor from Multiple Locations

Check from 2-3 regions to detect:

• Regional network issues
• Firewall misconfigurations
• ISP-specific problems

4. Use Appropriate Check Intervals

• Critical databases: 30-60 seconds
• Background services: 2-5 minutes
• SSH/Admin access: 5-15 minutes

5. Whitelist Monitoring IPs

If your firewall blocks connections, whitelist BlackTide check runner IPs:

• US East: Contact support for IP ranges
• EU West: Contact support for IP ranges

Troubleshooting

Check Fails with "Connection Timeout"

Possible Causes:

• Service is down
• Firewall blocking connections
• Network routing issue
• Port is closed

Solution:

1. Verify service is running:

   # Check if PostgreSQL is running
   sudo systemctl status postgresql
   
   # Check if port is listening
   sudo netstat -tlnp | grep 5432

2. Test connectivity manually:

   # Test TCP connection
   nc -zv db.example.com 5432
   
   # Or use telnet
   telnet db.example.com 5432

3. Check firewall rules:

   # List iptables rules
   sudo iptables -L -n
   
   # Check if port is allowed
   sudo ufw status | grep 5432

4. Whitelist BlackTide IPs in firewall

Check Fails with "Connection Refused"

Cause: Port is closed or service not listening.

Solution:

• Verify service is running
• Check service is bound to correct interface (0.0.0.0 for external access)
• Verify port number is correct

Intermittent Failures

Possible Causes:

• Service is overloaded
• Connection pool exhausted
• Network congestion

Solution:

• Increase connection pool size
• Scale service horizontally
• Increase check interval to reduce load

Limitations

TCP vs Application Health

TCP monitors only verify the port is open. They don't verify:

• Service is responding correctly
• Database queries succeed
• Authentication works
• Data is accessible

For comprehensive monitoring, combine TCP with HTTP health checks.

No Protocol-Specific Validation

TCP monitors don't send protocol-specific commands. They can't:

• Execute SQL queries
• Authenticate to SSH
• Send email via SMTP

Use Custom monitors for protocol-specific checks.

Next Steps

• HTTP Monitor: Monitor web endpoints
• ICMP Monitor: Basic network connectivity
• Configure Alert Rules for your TCP monitors
• Best Practices for effective monitoring