Routing: Multiple Upstreams Behind One Domain

Split one public API domain across several backends while keeping routing rules readable.

Split one public API domain across several backends while keeping routing rules readable.

Last reviewed: 2026-03-06

When to use this

Use this approach when you want a config-first Cloudflare Worker gateway behavior that is already implemented in the repository and covered by tests or canonical examples.

What this does not do

• It does not add unsupported product features such as rate limiting, caching, API keys, analytics, or OpenAPI generation.

• It does not replace upstream application logic that still belongs in your services.

• It does not remove the need to validate environment variables, bindings, and route intent before deploy.

Repo-grounded example

{
  "servers": [
    { "alias": "upstream", "url": "https://api.example.com/base" }
  ],
  "paths": [
    {
      "method": "GET",
      "path": "/proxy/{.+}",
      "integration": {
        "type": "http_proxy",
        "server": "upstream"
      }
    }
  ]
}

This example is grounded in the current implementation shape: JSON config, path-based routing, optional auth, request mapping, and Worker-native integrations.

Troubleshooting

• Confirm the route path and HTTP method match what the worker receives.

• Confirm the config source is the one you expect: local file, KV, or SAG_API_CONFIG_JSON.

• Confirm required bindings and environment variables are present before debugging downstream logic.

• Run the existing tests in the main gateway repo when a config change appears correct but runtime behavior disagrees.

Related docs

• README

• add and remove prefix

• gateway troubleshooting matrix