feat: Webhook retries

artifacts/flagger/crd.yaml

@@ -1132,6 +1132,9 @@ spec:
                             description: Request timeout for this webhook
                             type: string
                             pattern: "^[0-9]+(m|s)"
+                          retries:
+                            description: Number of retries for this webhook
+                            type: number
                           metadata:
                             description: Metadata (key-value pairs) for this webhook
                             type: object

charts/flagger/crds/crd.yaml

@@ -1132,6 +1132,9 @@ spec:
                             description: Request timeout for this webhook
                             type: string
                             pattern: "^[0-9]+(m|s)"
+                          retries:
+                            description: Number of retries for this webhook
+                            type: number
                           metadata:
                             description: Metadata (key-value pairs) for this webhook
                             type: object

docs/gitbook/usage/webhooks.md

@@ -41,6 +41,7 @@ Spec:
       - name: "start gate"
         type: confirm-rollout
         url: http://flagger-loadtester.test/gate/approve
+        retries: 5
       - name: "helm test"
         type: pre-rollout
         url: http://flagger-helmtester.flagger/
@@ -72,6 +73,7 @@ Spec:
       - name: "send to Slack"
         type: event
         url: http://event-recevier.notifications/slack
+        retries: 3
         metadata:
           environment: "test"
           cluster: "flagger-test"
@@ -122,6 +124,8 @@ Event payload (HTTP POST):
 The event receiver can create alerts based on the received phase 
 (possible values: `Initialized`, `Waiting`, `Progressing`, `Promoting`, `Finalising`, `Succeeded` or `Failed`).
 
+The webhook request can be retried by specifying a positive integer in the `retries` field.
+
 ## Load Testing
 
 For workloads that are not receiving constant traffic Flagger can be configured with a webhook,

go.mod

@@ -48,6 +48,8 @@ require (
 	github.com/google/uuid v1.3.1 // indirect
 	github.com/googleapis/enterprise-certificate-proxy v0.3.1 // indirect
 	github.com/h2non/parth v0.0.0-20190131123155-b4df798d6542 // indirect
+	github.com/hashicorp/go-cleanhttp v0.5.2 // indirect
+	github.com/hashicorp/go-retryablehttp v0.7.4 // indirect
 	github.com/imdario/mergo v0.3.15 // indirect
 	github.com/influxdata/line-protocol v0.0.0-20210922203350-b1ad95c89adf // indirect
 	github.com/jmespath/go-jmespath v0.4.0 // indirect

go.sum

@@ -112,6 +112,11 @@ github.com/googleapis/gax-go/v2 v2.12.0/go.mod h1:y+aIqrI5eb1YGMVJfuV3185Ts/D7qK
 github.com/grpc-ecosystem/grpc-gateway v1.16.0/go.mod h1:BDjrQk3hbvj6Nolgz8mAMFbcEtjT1g+wF4CSlocrBnw=
 github.com/h2non/parth v0.0.0-20190131123155-b4df798d6542 h1:2VTzZjLZBgl62/EtslCrtky5vbi9dd7HrQPQIx6wqiw=
 github.com/h2non/parth v0.0.0-20190131123155-b4df798d6542/go.mod h1:Ow0tF8D4Kplbc8s8sSb3V2oUCygFHVp8gC3Dn6U4MNI=
+github.com/hashicorp/go-cleanhttp v0.5.2 h1:035FKYIWjmULyFRBKPs8TBQoi0x6d9G4xc9neXJWAZQ=
+github.com/hashicorp/go-cleanhttp v0.5.2/go.mod h1:kO/YDlP8L1346E6Sodw+PrpBSV4/SoxCXGY6BqNFT48=
+github.com/hashicorp/go-hclog v0.9.2/go.mod h1:5CU+agLiy3J7N7QjHK5d05KxGsuXiQLrjA0H7acj2lQ=
+github.com/hashicorp/go-retryablehttp v0.7.4 h1:ZQgVdpTdAL7WpMIwLzCfbalOcSUdkDZnpUv3/+BxzFA=
+github.com/hashicorp/go-retryablehttp v0.7.4/go.mod h1:Jy/gPYAdjqffZ/yFGCFV2doI5wjtH1ewM9u8iYVjtX8=
 github.com/imdario/mergo v0.3.15 h1:M8XP7IuFNsqUx6VPK2P9OSmsYsI/YFaGil0uD21V3dM=
 github.com/imdario/mergo v0.3.15/go.mod h1:WBLT9ZmE3lPoWsEzCh9LPo3TiwVN+ZKEjmz+hD27ysY=
 github.com/influxdata/influxdb-client-go/v2 v2.12.3 h1:28nRlNMRIV4QbtIUvxhWqaxn0IpXeMSkY/uJa/O/vC4=
@@ -176,6 +181,7 @@ github.com/stoewer/go-strcase v1.2.0/go.mod h1:IBiWB2sKIp3wVVQ3Y035++gc+knqhUQag
 github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
 github.com/stretchr/objx v0.4.0/go.mod h1:YvHI0jy2hoMjB+UWwv71VJQ9isScKT/TqJzVSSt89Yw=
 github.com/stretchr/objx v0.5.0/go.mod h1:Yh+to48EsGEfYuaHDzXPcE3xhTkx73EhmCGUpEOglKo=
+github.com/stretchr/testify v1.2.2/go.mod h1:a8OnRcib4nhh0OaRAV+Yts87kKdq0PP7pXfy6kDkUVs=
 github.com/stretchr/testify v1.3.0/go.mod h1:M5WIy9Dh21IEIfnGCwXGc5bZfKNJtfHm1UVUgZn+9EI=
 github.com/stretchr/testify v1.5.1/go.mod h1:5W2xD1RspED5o8YsWQXVCued0rvSQ+mT+I5cxcmMvtA=
 github.com/stretchr/testify v1.7.0/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/h/Wwjteg=

kustomize/base/flagger/crd.yaml

@@ -1132,6 +1132,9 @@ spec:
                             description: Request timeout for this webhook
                             type: string
                             pattern: "^[0-9]+(m|s)"
+                          retries:
+                            description: Number of retries for this webhook
+                            type: number
                           metadata:
                             description: Metadata (key-value pairs) for this webhook
                             type: object

pkg/apis/flagger/v1beta1/canary.go

@@ -394,6 +394,10 @@ type CanaryWebhook struct {
 	// Metadata (key-value pairs) for this webhook
 	// +optional
 	Metadata *map[string]string `json:"metadata,omitempty"`
+
+	// Number of retries for this webhook
+	// +optional
+	Retries int `json:"retries,omitempty"`
 }
 
 // CanaryWebhookPayload holds the deployment info and metadata sent to webhooks

pkg/controller/webhook.go

@@ -18,21 +18,21 @@ package controller
 
 import (
 	"bytes"
-	"context"
 	"encoding/json"
 	"errors"
 	"fmt"
 	"io"
-	"net/http"
 	"net/url"
 	"strconv"
 	"time"
 
+	"github.com/hashicorp/go-retryablehttp"
+
 	flaggerv1 "github.com/fluxcd/flagger/pkg/apis/flagger/v1beta1"
 	"github.com/fluxcd/flagger/pkg/canary"
 )
 
-func callWebhook(webhook string, payload interface{}, timeout string) error {
+func callWebhook(webhook string, payload interface{}, timeout string, retries int) error {
 	payloadBin, err := json.Marshal(payload)
 	if err != nil {
 		return err
@@ -43,7 +43,11 @@ func callWebhook(webhook string, payload interface{}, timeout string) error {
 		return err
 	}
 
-	req, err := http.NewRequest("POST", hook.String(), bytes.NewBuffer(payloadBin))
+	httpClient := retryablehttp.NewClient()
+	httpClient.RetryMax = retries
+	httpClient.Logger = nil
+
+	req, err := retryablehttp.NewRequest("POST", hook.String(), bytes.NewBuffer(payloadBin))
 	if err != nil {
 		return err
 	}
@@ -53,16 +57,14 @@ func callWebhook(webhook string, payload interface{}, timeout string) error {
 	if timeout == "" {
 		timeout = "10s"
 	}
-
 	t, err := time.ParseDuration(timeout)
 	if err != nil {
 		return err
 	}
 
-	ctx, cancel := context.WithTimeout(req.Context(), t)
-	defer cancel()
+	httpClient.HTTPClient.Timeout = t
 
-	r, err := http.DefaultClient.Do(req.WithContext(ctx))
+	r, err := httpClient.Do(req)
 	if err != nil {
 		return err
 	}
@@ -98,7 +100,7 @@ func CallWebhook(canary flaggerv1.Canary, phase flaggerv1.CanaryPhase, w flagger
 		w.Timeout = "10s"
 	}
 
-	return callWebhook(w.URL, payload, w.Timeout)
+	return callWebhook(w.URL, payload, w.Timeout, w.Retries)
 }
 
 func CallEventWebhook(r *flaggerv1.Canary, w flaggerv1.CanaryWebhook, message, eventtype string) error {
@@ -124,7 +126,7 @@ func CallEventWebhook(r *flaggerv1.Canary, w flaggerv1.CanaryWebhook, message, e
 			payload.Metadata[key] = value
 		}
 	}
-	return callWebhook(w.URL, payload, "5s")
+	return callWebhook(w.URL, payload, "5s", w.Retries)
 }
 
 func canaryChecksum(c flaggerv1.Canary) string {

pkg/controller/webhook_test.go

@@ -263,3 +263,29 @@ func TestCanaryChecksum(t *testing.T) {
 	require.NotEqual(t, canary3sum, canary1sum)
 	require.NotEqual(t, canary4sum, canary1sum)
 }
+
+func TestCallWebhook_Retries(t *testing.T) {
+	retries := 1
+	failures := 0
+	ts := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		if failures <= retries-1 {
+			w.WriteHeader(http.StatusInternalServerError)
+			failures++
+		} else {
+			w.WriteHeader(http.StatusAccepted)
+		}
+	}))
+	defer ts.Close()
+	hook := flaggerv1.CanaryWebhook{
+		Name:    "validation",
+		URL:     ts.URL,
+		Retries: retries,
+	}
+
+	err := CallWebhook(
+		flaggerv1.Canary{
+			ObjectMeta: metav1.ObjectMeta{
+				Name: "podinfo", Namespace: corev1.NamespaceDefault}},
+		flaggerv1.CanaryPhaseProgressing, hook)
+	require.NoError(t, err)
+}