Merge pull request #2268 from O-sura/apip-resiliency

Adding end-to-end Timeout Functionality

Author: Krishanx92

gateway/build-manifest.yaml

@@ -61,7 +61,7 @@ policies:
       version: v1.0.2
       gomodule: github.com/wso2/gateway-controllers/policies/mcp-acl-list@v1
     - name: mcp-auth
-      version: v1.0.3
+      version: v1.1.0
       gomodule: github.com/wso2/gateway-controllers/policies/mcp-auth@v1
     - name: mcp-authz
       version: v1.0.1

gateway/configs/config-template.toml

@@ -212,6 +212,18 @@ connect_timeout_ms = 5000
 server_header_transformation = "OVERWRITE"
 server_header_value = "WSO2 API Platform"
 
+# HTTP Connection Manager (downstream) timeouts
+# A value of zero disables any of these timeout settings.
+[router.http_listener.timeouts]
+# Max duration for the entire downstream request
+request_timeout = "0s"
+# Max duration to receive the complete request headers
+request_headers_timeout = "0s"
+# Idle timeout for a single HTTP stream/request
+stream_idle_timeout = "5m"
+# Idle timeout for the downstream connection
+idle_timeout = "1h"
+
 [router.policy_engine]
 host = "policy-engine"
 port = 9001

gateway/gateway-controller/api/management-openapi.yaml

@@ -4194,6 +4194,8 @@ components:
           description: List of API-level policies applied to all operations unless overridden
           items:
             $ref: "#/components/schemas/Policy"
+        resilience:
+          $ref: "#/components/schemas/Resilience"
         operations:
           type: array
           description: List of HTTP operations/routes
@@ -4223,7 +4225,8 @@ components:
           example: my-upstream-1
         basePath:
           type: string
-          description: Base path prefix for all endpoints in this upstream (e.g., /api/v2). All requests to this upstream will have this path prepended.
+          description: Base path prefix for all endpoints in this upstream (e.g., /api/v2). All requests to this upstream will have this path prepended. Must start with '/' and must not end with '/'; omit for root.
+          pattern: '^/[a-zA-Z0-9\-._~!$&''()*+,;=:@%/]*[^/]$'
           example: /api/v2
         timeout:
           $ref: "#/components/schemas/UpstreamTimeout"
@@ -4243,7 +4246,7 @@ components:
                 example: http://prod-backend-1:5000
               weight:
                 type: integer
-                description: Weight for load balancing (optional, default 100)
+                description: Relative weight for load balancing across multiple upstream targets. Reserved for future multi-target load balancing; not applied yet (only the first target is currently used).
                 minimum: 0
                 maximum: 100
                 example: 80
@@ -4258,6 +4261,25 @@ components:
           pattern: '^\d+(\.\d+)?(ms|s|m|h)$'
           example: 5s
 
+    Resilience:
+      type: object
+      description: >
+        Backend/route timeout configuration. Maps to Envoy RouteAction timeouts.
+        Can be set at the API level (applies to all routes) and/or the operation level
+        (applies to that operation's route). When set at both levels, the operation-level
+        value takes precedence. When unset, the gateway's global route timeout defaults apply.
+      properties:
+        timeout:
+          type: string
+          description: Maximum time for the entire route (request to upstream response). "0s" disables the timeout.
+          pattern: '^\d+(\.\d+)?(ms|s|m|h)$'
+          example: 15s
+        idleTimeout:
+          type: string
+          description: Per-route stream idle timeout (overrides the listener stream idle timeout for this route). "0s" disables the timeout.
+          pattern: '^\d+(\.\d+)?(ms|s|m|h)$'
+          example: 0s
+
     Upstream:
       type: object
       oneOf:
@@ -4313,6 +4335,8 @@ components:
           description: List of policies applied only to this operation (overrides or adds to API-level policies)
           items:
             $ref: "#/components/schemas/Policy"
+        resilience:
+          $ref: "#/components/schemas/Resilience"
 
     Policy:
       type: object
@@ -5322,6 +5346,11 @@ components:
           minLength: 1
           maxLength: 253
           example: "mcp1.example.com"
+        upstreamDefinitions:
+          type: array
+          description: List of reusable upstream definitions with optional timeout configurations. Referenced by upstream.ref.
+          items:
+            $ref: "#/components/schemas/UpstreamDefinition"
         upstream:
           description: The backend MCP server url and auth configurations
           allOf:
@@ -5350,6 +5379,15 @@ components:
           enum: [deployed, undeployed]
           default: deployed
           example: deployed
+        resilience:
+          $ref: '#/components/schemas/Resilience'
+          description: >
+            API-level backend/route timeout configuration. Applies to the traffic-forwarding
+            routes generated for this MCP proxy (GET/POST/DELETE on the MCP resource path).
+            Supported at the API level only. Because MCP transports are long-lived streams,
+            the route timeout defaults to disabled ("0s") for MCP unless a timeout is set here
+            (unlike REST/LLM, which fall back to the gateway's global route timeout); the idle
+            timeout remains the liveness guard.
 
     MCPTool:
       type: object
@@ -5790,6 +5828,11 @@ components:
           type: string
           description: Template name to use for this LLM Provider
           example: openai
+        upstreamDefinitions:
+          type: array
+          description: List of reusable upstream definitions with optional timeout configurations. Referenced by upstream.ref.
+          items:
+            $ref: "#/components/schemas/UpstreamDefinition"
         upstream:
           allOf:
             - $ref: "#/components/schemas/Upstream"
@@ -5818,6 +5861,13 @@ components:
           enum: [deployed, undeployed]
           default: deployed
           example: deployed
+        resilience:
+          $ref: '#/components/schemas/Resilience'
+          description: >
+            API-level backend/route timeout configuration. Applies to all routes generated
+            for this LLM Provider (the routes that forward traffic upstream). Supported at the
+            API level only - LLM routes are synthesized by the gateway, so there is no
+            operation-level override.
 
     UpstreamAuth:
       type: object
@@ -6108,6 +6158,13 @@ components:
           enum: [deployed, undeployed]
           default: deployed
           example: deployed
+        resilience:
+          $ref: '#/components/schemas/Resilience'
+          description: >
+            API-level backend/route timeout configuration. Applies to all routes generated
+            for this LLM Proxy (the routes that forward traffic upstream). Supported at the
+            API level only - LLM routes are synthesized by the gateway, so there is no
+            operation-level override.
 
     SecretConfigurationRequest:
       type: object