Put response cache in GA and remove preview_ prefix (#8678)

Signed-off-by: Benjamin <[email protected]>

Author: bnjjj

.changesets/feat_response_cache_ga.md

@@ -0,0 +1,83 @@
+### Response caching is now Generally Available 🎉 ([PR #8678](https://github.com/apollographql/router/pull/8678))
+
+We're excited to announce that **response caching is now Generally Available (GA)** and ready for production use!
+
+Response caching is a powerful feature that enables the router to cache subgraph query responses using Redis, dramatically improving query latency and reducing load on your underlying services. Unlike traditional HTTP caching solutions, response caching provides GraphQL-aware caching at the entity and root field level, making cached data reusable across different users and queries.
+
+#### Key benefits
+
+- **Improved performance**: Cache origin responses and reuse them across queries to reduce latency
+- **Reduced subgraph load**: Minimize redundant requests to your subgraphs by serving cached data
+- **Entity-level caching**: Cache individual entity representations independently, enabling fine-grained control over data freshness
+- **Flexible cache control**: Set different TTLs for different types of data based on `@cacheControl` directives in your schema or the `Cache-Control` response header
+- **Privacy-aware**: Mix public and private data safely—share cached data across users while maintaining privacy for personalized data
+- **GraphQL-native**: Solve unique GraphQL caching challenges that traditional CDNs can't address, such as mixed TTLs and high data duplication
+- **Active cache invalidation**: Use the `@cacheTag` directive in your schema to tag cached data, then actively invalidate specific cache entries via an HTTP endpoint when data changes
+
+#### What's cached
+
+The router caches two kinds of data:
+- **Root query fields**: Cached as complete units (the entire response for these root fields)
+- **Entity representations**: Cached independently—each origin's contribution to an entity is cached separately and can be reused across different queries
+
+
+#### Configuration
+
+Response caching uses Redis as the cache backend and can be configured per subgraph with invalidation support:
+
+```yaml
+response_cache:
+  enabled: true
+  # Configure the invalidation endpoint
+  invalidation:
+    listen: 127.0.0.1:4000
+    path: /invalidation
+  subgraph:
+    all:
+      enabled: true
+      ttl: 30s
+      redis:
+        urls:
+          - redis://127.0.0.1:6379
+      invalidation:
+        enabled: true
+        shared_key: ${env.INVALIDATION_SHARED_KEY}
+```
+
+The router determines cache TTLs from `Cache-Control` HTTP headers returned by your origins. You also get comprehensive Redis metrics to monitor cache performance, including connection health, command execution, and operational insights.
+
+#### Cache invalidation with @cacheTag
+
+Tag your cached data using the `@cacheTag` directive in your subgraph schema, then actively invalidate specific cache entries when data changes:
+
+```graphql
+type Query {
+  user(id: ID!): User @cacheTag(format: "user-{$args.id}")
+  users: [User!]! @cacheTag(format: "users-list")
+}
+
+type User @key(fields: "id") @cacheTag(format: "user-{$key.id}") {
+  id: ID!
+  name: String!
+}
+```
+
+Send invalidation requests to remove cached data before TTL expires:
+
+```bash
+curl --request POST \
+	--header "authorization: $INVALIDATION_SHARED_KEY" \
+	--header 'content-type: application/json' \
+	--url http://localhost:4000/invalidation \
+	--data '[{"kind":"cache_tag","subgraphs":["posts"],"cache_tag":"user-42"}]'
+```
+
+#### Additional features
+
+- **Cache debugger**: See exactly what's being cached during development with detailed cache key inspection
+- **Redis cluster support**: Scale your cache with Redis cluster deployments and benefit from read replicas for improved performance and availability
+- **Comprehensive metrics**: Monitor cache performance with detailed Redis-specific metrics including connection health, command execution, and latency
+
+For complete documentation, configuration options, and best practices, see the [response caching documentation](/router/routing/performance/caching/response-caching/overview).
+
+By [@bnjjj](https://github.com/bnjjj) in https://github.com/apollographql/router/pull/8678

.changesets/fix_response_cache_invalidation_endpoint.md

@@ -1,15 +1,15 @@
 ### Enable invalidation endpoint for response cache when any subgraph has invalidation enabled ([PR #8680](https://github.com/apollographql/router/pull/8680))
 
-Previously, the response cache invalidation endpoint was only enabled when global invalidation was enabled via `preview_response_cache.subgraph.all.invalidation.enabled`. This meant that if you enabled invalidation for only specific subgraphs without enabling it globally, the invalidation endpoint would not be started, preventing cache invalidation requests from being processed.
+Previously, the response cache invalidation endpoint was only enabled when global invalidation was enabled via `response_cache.subgraph.all.invalidation.enabled`. This meant that if you enabled invalidation for only specific subgraphs without enabling it globally, the invalidation endpoint would not be started, preventing cache invalidation requests from being processed.
 
 Now, the invalidation endpoint is enabled if either:
-- Global invalidation is enabled (`preview_response_cache.subgraph.all.invalidation.enabled: true`), OR
+- Global invalidation is enabled (`response_cache.subgraph.all.invalidation.enabled: true`), OR
 - Any individual subgraph has invalidation enabled
 
 This allows for more flexible configuration where you can enable invalidation selectively for specific subgraphs:
 
 ```yaml
-preview_response_cache:
+response_cache:
   enabled: true
   invalidation:
     listen: 127.0.0.1:4000

.circleci/config.yml

@@ -1161,4 +1161,4 @@ workflows:
             - secops-oidc
             - github-orb
           git-base-revision: <<#pipeline.git.base_revision>><<pipeline.git.base_revision>><</pipeline.git.base_revision >>
-          disabled-signatures: "rules.providers.semgrep.security.javascript.lang.security.detect-insecure-websocket"
+          disabled-signatures: "rules.providers.semgrep.security.javascript.lang.security.detect-insecure-websocket"

DEVELOPMENT.md

@@ -184,4 +184,4 @@ For full information on available options, including HTML reports and `lcov.info
 
 ## Project maintainers
 
-Apollo Graph, Inc.
+Apollo Graph, Inc.

apollo-router/Cargo.toml

@@ -98,8 +98,8 @@ fred = { version = "10.1.0", features = [
     "i-cluster",
     "tcp-user-timeouts",
     "metrics",
-    "replicas",
-    "serde-json"
+    "serde-json",
+    "replicas"
 ] }
 futures = { version = "0.3.30", features = ["thread-pool"] }
 graphql_client = "0.14.0"
@@ -310,7 +310,7 @@ fred = { version = "10.1.0", features = [
     "mocks",
     "i-cluster",
     "tcp-user-timeouts",
-    "replicas"
+    "replicas",
 ] }
 futures-test = "0.3.30"
 insta.workspace = true

apollo-router/src/configuration/metrics.rs

@@ -323,7 +323,7 @@ impl InstrumentData {
 
         populate_config_instrument!(
             apollo.router.config.response_cache,
-            "$.preview_response_cache",
+            "$.response_cache",
             opt.enabled,
             "$[?(@.enabled)]",
             opt.debug,

apollo-router/src/configuration/migrations/2044-response_cache_ga.yaml

@@ -0,0 +1,5 @@
+description: Response cache feature is now GA
+actions:
+  - type: move
+    from: preview_response_cache
+    to: response_cache

apollo-router/src/configuration/mod.rs

@@ -586,12 +586,12 @@ impl Configuration {
         }
 
         // response & entity caching
-        if self.apollo_plugin_enabled("preview_response_cache")
+        if self.apollo_plugin_enabled("response_cache")
             && self.apollo_plugin_enabled("preview_entity_cache")
         {
             return Err(ConfigurationError::InvalidConfiguration {
                 message: "entity cache and response cache features are mutually exclusive",
-                error: "either set preview_response_cache.enabled: false or preview_entity_cache.enabled: false in your router yaml configuration".into(),
+                error: "either set response_cache.enabled: false or preview_entity_cache.enabled: false in your router yaml configuration".into(),
             });
         }
 

apollo-router/src/configuration/snapshots/apollo_router__configuration__tests__schema_generation.snap

@@ -1929,6 +1929,10 @@ expression: "&schema"
       "type": "object"
     },
     "Config7": {
+      "description": "Configuration for the progressive override plugin",
+      "type": "object"
+    },
+    "Config8": {
       "additionalProperties": false,
       "description": "Configuration for response caching",
       "properties": {
@@ -1974,7 +1978,7 @@ expression: "&schema"
       ],
       "type": "object"
     },
-    "Config8": {
+    "Config9": {
       "additionalProperties": false,
       "description": "Redis cache configuration",
       "properties": {
@@ -2092,10 +2096,6 @@ expression: "&schema"
       ],
       "type": "object"
     },
-    "Config9": {
-      "description": "Configuration for the progressive override plugin",
-      "type": "object"
-    },
     "ConnectorConfiguration": {
       "properties": {
         "all": {
@@ -9245,7 +9245,7 @@ expression: "&schema"
         "redis": {
           "anyOf": [
             {
-              "$ref": "#/definitions/Config8"
+              "$ref": "#/definitions/Config9"
             },
             {
               "type": "null"
@@ -11947,8 +11947,8 @@ expression: "&schema"
       },
       "type": "object"
     },
-    "^preview_response_cache$": {
-      "$ref": "#/definitions/Config7"
+    "^response_cache$": {
+      "$ref": "#/definitions/Config8"
     }
   },
   "properties": {
@@ -12144,7 +12144,7 @@ expression: "&schema"
       "$ref": "#/definitions/FileUploadsConfig"
     },
     "progressive_override": {
-      "$ref": "#/definitions/Config9"
+      "$ref": "#/definitions/Config7"
     },
     "rhai": {
       "$ref": "#/definitions/RhaiConfig"

apollo-router/src/configuration/snapshots/apollo_router__configuration__tests__upgrade_old_configuration@response_cache_preview.yaml.snap

@@ -0,0 +1,25 @@
+---
+source: apollo-router/src/configuration/tests.rs
+expression: new_config
+---
+---
+response_cache:
+  enabled: true
+  subgraph:
+    all:
+      redis:
+        urls:
+          - "redis://test"
+        required_to_start: true
+      enabled: true
+      invalidation:
+        enabled: true
+        shared_key: invalidate
+    subgraphs:
+      accounts:
+        enabled: false
+      products:
+        ttl: 120s
+        invalidation:
+          enabled: true
+          shared_key: invalidate