fix(retry): fix the retry policy do not take affect

.github/workflows/tests.yml

@@ -6,8 +6,8 @@ jobs:
   unit-benchmark-test:
     strategy:
       matrix:
-        go: [ 1.17, 1.18, 1.19 ]
-        os: [ X64, ARM64 ]
+        go: [ "1.18", "1.19", "1.20", "1.21", "1.22"]
+        os: [ X64 ]
     runs-on: ${{ matrix.os }}
     steps:
       - uses: actions/checkout@v3

README.md

@@ -31,9 +31,10 @@ To enable xDS mode in Kitex, we should invoke `xds.Init()` to initialize the xds
 #### Bootstrap
 The xdsClient is responsible for the interaction with the xDS Server (i.e. Istio). It needs some environment variables for initialization, which need to be set inside the `spec.containers.env` of the Kubernetes Manifest file in YAML format.
 
-* `POD_NAMESPACE`: the namespace of the current service.
+*  `POD_NAMESPACE`: the namespace of the current service.
 *  `POD_NAME`: the name of this pod.
 *  `INSTANCE_IP`: the ip of this pod.
+*  `KITEX_XDS_METAS`: the metadata of this xDS node.
 
 Add the following part to the definition of your container that uses xDS-enabled Kitex client.
 
@@ -50,6 +51,8 @@ valueFrom:
 valueFrom:
   fieldRef:
     fieldPath: status.podIP
+- name: KITEX_XDS_METAS
+  value: '{"CLUSTER_ID":"Kubernetes","DNS_AUTO_ALLOCATE":"true","DNS_CAPTURE":"true","INSTANCE_IPS":"$(INSTANCE_IP)","NAMESPACE":"$(POD_NAMESPACE)"}'
 ```
 
 ### Client-side
@@ -231,6 +234,10 @@ spec:
           failure_percentage_threshold: 10
           # the failure percentage request volume
           failure_percentage_request_volume: 101
+  workloadSelector:
+    labels:
+      # the label of the client pod.
+      app.kubernetes.io/name: kitex-client
 ```
 
 #### RateLimit
@@ -262,12 +269,72 @@ spec:
               max_tokens: 4
   workloadSelector:
     labels:
-      # the label of the service pod.
+      # the label of the server pod.
       app.kubernetes.io/name: kitex-server
 
 ```
 
+#### Retry
+
+Support using VirtualService and EnvoyFilter to config retry policy, the EnvoyFilter has more configuration.
 
+```
+apiVersion: networking.istio.io/v1
+kind: VirtualService
+metadata:
+  name: retry-sample
+  namespace: default
+spec:
+  hosts:
+  - hello.prod.svc.cluster.local:21001
+  http:
+  - route:
+    - destination:
+        host: hello.prod.svc.cluster.local:21001
+    retries:
+      attempts: 1
+      perTryTimeout: 2s
+```
+
+```
+apiVersion: networking.istio.io/v1alpha3
+kind: EnvoyFilter
+metadata:
+  name: retry-enhance
+  namespace: default
+spec:
+  configPatches:
+  - applyTo: HTTP_ROUTE
+    match:
+      context: SIDECAR_OUTBOUND
+      routeConfiguration:
+        # service name, should obey FQDN
+        name: hello.default.svc.cluster.local:21001
+        vhost: 
+          # service name, should obey FQDN
+          name: hello.default.svc.cluster.local:21001
+    patch:
+      operation: MERGE
+      value:
+        route:
+          retryPolicy:
+            numRetries: 3
+            perTryTimeout: 100ms
+            retryBackOff:
+              baseInterval: 100ms
+              maxInterval: 100ms
+            retriableHeaders:
+              - name: "kitexRetryErrorRate"
+                stringMatch:
+                  exact: "0.29"
+              - name: "kitexRetryMethods"
+                stringMatch:
+                  exact: "Echo,Greet"
+  workloadSelector:
+    labels:
+      # the label of the service pod.
+      app.kubernetes.io/name: kitex-client
+```
 
 ## Example
 The usage is as follows:

README_CN.md

@@ -32,9 +32,10 @@ Kitex 通过外部扩展 [kitex-contrib/xds](https://github.com/kitex-contrib/xd
 xdsClient 负责与控制面（例如 Istio）交互，以获得所需的 xDS 资源。在初始化时，需要读取环境变量用于构建 node 标识。所以，需要在K8S 的容器配置文件 `spec.containers.env` 部分加入以下几个环境变量。
 
 
-* `POD_NAMESPACE`: 当前 pod 所在的 namespace。
+*  `POD_NAMESPACE`: 当前 pod 所在的 namespace。
 *  `POD_NAME`: pod 名。
 *  `INSTANCE_IP`: pod 的 ip。
+*  `KITEX_XDS_METAS`: 用于构建 node 标识的元信息，格式为 json 字符串。
 
 在需要使用 xDS 功能的容器配置中加入以下定义即可：
 
@@ -51,6 +52,8 @@ valueFrom:
 valueFrom:
   fieldRef:
     fieldPath: status.podIP
+- name: KITEX_XDS_METAS
+  value: '{"CLUSTER_ID":"Kubernetes","DNS_AUTO_ALLOCATE":"true","DNS_CAPTURE":"true","INSTANCE_IPS":"$(INSTANCE_IP)","NAMESPACE":"$(POD_NAMESPACE)"}'
 ```
 
 ### Kitex 客户端
@@ -221,6 +224,10 @@ spec:
           failure_percentage_threshold: 10
           # 触发熔断请求量
           failure_percentage_request_volume: 101
+  workloadSelector:
+    labels:
+      # the label of the client pod.
+      app.kubernetes.io/name: kitex-client
 ```
 
 #### 限流配置
@@ -256,6 +263,67 @@ spec:
       app.kubernetes.io/name: kitex-server
 
 ```
+#### 重试配置
+
+重试支持两个配置方式：VirtualService 和 EnvoyFilter，EnvoyFilter 支持更丰富的重试策略。
+
+```
+apiVersion: networking.istio.io/v1
+kind: VirtualService
+metadata:
+  name: retry-sample
+  namespace: default
+spec:
+  hosts:
+  - hello.prod.svc.cluster.local:21001
+  http:
+  - route:
+    - destination:
+        host: hello.prod.svc.cluster.local:21001
+    retries:
+      attempts: 1
+      perTryTimeout: 2s
+```
+
+```
+apiVersion: networking.istio.io/v1alpha3
+kind: EnvoyFilter
+metadata:
+  name: retry-enhance
+  namespace: default
+spec:
+  configPatches:
+  - applyTo: HTTP_ROUTE
+    match:
+      context: SIDECAR_OUTBOUND
+      routeConfiguration:
+        # service name, should obey FQDN
+        name: hello.default.svc.cluster.local:21001
+        vhost: 
+          # service name, should obey FQDN
+          name: hello.default.svc.cluster.local:21001
+    patch:
+      operation: MERGE
+      value:
+        route:
+          retryPolicy:
+            numRetries: 3
+            perTryTimeout: 100ms
+            retryBackOff:
+              baseInterval: 100ms
+              maxInterval: 100ms
+            retriableHeaders:
+              - name: "kitexRetryErrorRate"
+                stringMatch:
+                  exact: "0.29"
+              - name: "kitexRetryMethods"
+                stringMatch:
+                  exact: "Echo,Greet"
+  workloadSelector:
+    labels:
+      # the label of the service pod.
+      app.kubernetes.io/name: kitex-client
+```
 
 ## 示例
 完整的客户端用法如下:

core/xdsresource/rds.go

@@ -19,13 +19,21 @@ package xdsresource
 import (
 	"encoding/json"
 	"fmt"
+	"strconv"
+	"strings"
 	"time"
 
 	v3routepb "github.com/envoyproxy/go-control-plane/envoy/config/route/v3"
 	"github.com/golang/protobuf/ptypes/any"
 	"google.golang.org/protobuf/proto"
 )
 
+const (
+	// If the error rate large than it, the retry policy do not take affect.
+	cBErrorRateKey = "kitexRetryErrorRate"
+	retryMethods   = "kitexRetryMethods"
+)
+
 // RouteConfigResource is used for routing
 // HTTPRouteConfig is the native http route config, which consists of a list of virtual hosts.
 // ThriftRouteConfig is converted from the routeConfiguration in thrift proxy, which can only be configured in the listener filter
@@ -34,6 +42,7 @@ type RouteConfigResource struct {
 	HTTPRouteConfig   *HTTPRouteConfig
 	ThriftRouteConfig *ThriftRouteConfig
 	MaxTokens         uint32
+	TokensPerFill     uint32
 }
 
 type HTTPRouteConfig struct {
@@ -64,7 +73,9 @@ type RetryPolicy struct {
 	NumRetries        int
 	PerTryTimeout     time.Duration
 	PerTryIdleTimeout time.Duration
+	CBErrorRate       float64
 	RetryBackOff      *RetryBackOff
+	Methods           []string
 }
 
 type Route struct {
@@ -181,6 +192,26 @@ func unmarshalRoutes(rs []*v3routepb.Route) ([]*Route, error) {
 					PerTryTimeout:     retryPolicy.GetPerTryTimeout().AsDuration(),
 					PerTryIdleTimeout: retryPolicy.GetPerTryIdleTimeout().AsDuration(),
 				}
+				// used for config the errRate.
+				for _, header := range retryPolicy.GetRetriableHeaders() {
+					match := header.GetStringMatch()
+					if match == nil {
+						continue
+					}
+					value := match.GetExact()
+					if value == "" {
+						continue
+					}
+					switch header.Name {
+					case cBErrorRateKey:
+						errRate, err := strconv.ParseFloat(value, 64)
+						if err == nil {
+							route.RetryPolicy.CBErrorRate = errRate
+						}
+					case retryMethods:
+						route.RetryPolicy.Methods = strings.Split(value, ",")
+					}
+				}
 				if backoff := retryPolicy.GetRetryBackOff(); backoff != nil {
 					route.RetryPolicy.RetryBackOff = &RetryBackOff{
 						BaseInterval: backoff.GetMaxInterval().AsDuration(),

go.mod

@@ -1,16 +1,16 @@
 module github.com/kitex-contrib/xds
 
-go 1.17
+go 1.18
 
 require (
-	github.com/bytedance/gopkg v0.1.0
+	github.com/bytedance/gopkg v0.1.1
 	github.com/cenkalti/backoff/v4 v4.1.0
 	github.com/cloudwego/kitex v0.10.3
 	github.com/cncf/udpa/go v0.0.0-20220112060539-c52dc94e7fbe
 	github.com/envoyproxy/go-control-plane v0.11.1
 	github.com/golang/protobuf v1.5.3
 	github.com/google/go-cmp v0.5.9
-	github.com/stretchr/testify v1.8.3
+	github.com/stretchr/testify v1.9.0
 	go.uber.org/atomic v1.11.0
 	google.golang.org/genproto/googleapis/rpc v0.0.0-20230526203410-71b5a4ffd15e
 	google.golang.org/protobuf v1.30.0
@@ -52,10 +52,10 @@ require (
 	github.com/tidwall/pretty v1.2.0 // indirect
 	github.com/twitchyliquid64/golang-asm v0.15.1 // indirect
 	golang.org/x/arch v0.2.0 // indirect
-	golang.org/x/net v0.17.0 // indirect
-	golang.org/x/sync v0.1.0 // indirect
-	golang.org/x/sys v0.13.0 // indirect
-	golang.org/x/text v0.13.0 // indirect
+	golang.org/x/net v0.24.0 // indirect
+	golang.org/x/sync v0.8.0 // indirect
+	golang.org/x/sys v0.19.0 // indirect
+	golang.org/x/text v0.14.0 // indirect
 	google.golang.org/genproto v0.0.0-20230526203410-71b5a4ffd15e // indirect
 	google.golang.org/genproto/googleapis/api v0.0.0-20230526203410-71b5a4ffd15e // indirect
 	google.golang.org/grpc v1.56.3 // indirect