Skip to main content

TCP Routing

TCP routing allows an API Gateway to handle raw network traffic at the transport layer, enabling it to serve as an entry point for non-HTTP protocols such as databases, game servers, messaging brokers, or custom protocols.

Proper TCP routing makes it possible to:

🔁 Consolidate entry points for multiple TCP-based services.
📡 Expose internal services to the internet with controlled access.
🔌 Handle any binary or custom protocol without deep inspection.

🔍 What are the Benefits of TCP Routing?​

Unlike HTTP and TLS routing, TCP routing operates at Layer 4 of the OSI model. The API Gateway uses the destination port or IP to determine the target backend—ideal for workloads that don't use HTTP.

With TCP routing, the gateway can route connections using:

  • Port-based rules (e.g., port 5432 → PostgreSQL service, port 4222 → NATS server).
  • Static or dynamic IP/port mappings to backend services.
  • Connection-level metadata, such as source IP or client certificate, if applicable.

Key Benefits:

  • Protocol Agnostic Routing: Support non-HTTP services or custom TCP protocols.
  • High Performance: Route traffic with minimal latency and overhead.
  • Secure Exposure of Internal Services: Control who can access backend services on the transport layer.

TCP Routing Examples​

The following examples showcase how you can route TCP traffic to your upstream services.

You must first create a TCP Address. When you create a TCP address, a random hostname and port will be assigned to you, e.g. 1.tcp.ngrok.io:12345. See the TCP Endpoints page for more details on how ngrok handles TCP endpoints.

Loading…