for aggregated apiserver availability, try multiple endpoints in parallel

staging/src/k8s.io/kube-aggregator/pkg/controllers/status/available_controller.go

@@ -242,33 +242,56 @@ func (c *AvailableConditionController) sync(key string) error {
 	}
 	// actually try to hit the discovery endpoint when it isn't local and when we're routing as a service.
 	if apiService.Spec.Service != nil && c.serviceResolver != nil {
-		discoveryURL, err := c.serviceResolver.ResolveEndpoint(apiService.Spec.Service.Namespace, apiService.Spec.Service.Name, apiService.Spec.Service.Port)
-		if err != nil {
-			return err
-		}
+		attempts := 5
+		results := make(chan error, attempts)
+		for i := 0; i < attempts; i++ {
+			go func() {
+				discoveryURL, err := c.serviceResolver.ResolveEndpoint(apiService.Spec.Service.Namespace, apiService.Spec.Service.Name, apiService.Spec.Service.Port)
+				if err != nil {
+					results <- err
+					return
+				}
 
-		errCh := make(chan error)
-		go func() {
-			resp, err := c.discoveryClient.Get(discoveryURL.String())
-			if resp != nil {
-				resp.Body.Close()
-			}
-			errCh <- err
-		}()
+				errCh := make(chan error)
+				go func() {
+					resp, err := c.discoveryClient.Get(discoveryURL.String())
+					if resp != nil {
+						resp.Body.Close()
+					}
+					errCh <- err
+				}()
+
+				select {
+				case err = <-errCh:
+					if err != nil {
+						results <- fmt.Errorf("no response from %v: %v", discoveryURL, err)
+						return
+					}
+
+					// we had trouble with slow dial and DNS responses causing us to wait too long.
+					// we added this as insurance
+				case <-time.After(6 * time.Second):
+					results <- fmt.Errorf("timed out waiting for %v", discoveryURL)
+					return
+				}
 
-		select {
-		case err = <-errCh:
+				results <- nil
+			}()
+		}
 
-		// we had trouble with slow dial and DNS responses causing us to wait too long.
-		// we added this as insurance
-		case <-time.After(6 * time.Second):
-			err = fmt.Errorf("timed out waiting for %v", discoveryURL)
+		var lastError error
+		for i := 0; i < attempts; i++ {
+			lastError = <-results
+			// if we had at least one success, we are successful overall and we can return now
+			if lastError == nil {
+				break
+			}
 		}
 
-		if err != nil {
+		if lastError != nil {
 			availableCondition.Status = apiregistration.ConditionFalse
 			availableCondition.Reason = "FailedDiscoveryCheck"
-			availableCondition.Message = fmt.Sprintf("no response from %v: %v", discoveryURL, err)
+			availableCondition.Message = lastError.Error()
 			apiregistration.SetAPIServiceCondition(apiService, availableCondition)
 			_, updateErr := updateAPIServiceStatus(c.apiServiceClient, originalAPIService, apiService)
 			if updateErr != nil {
@@ -276,7 +299,7 @@ func (c *AvailableConditionController) sync(key string) error {
 			}
 			// force a requeue to make it very obvious that this will be retried at some point in the future
 			// along with other requeues done via service change, endpoint change, and resync
-			return err
+			return lastError
 		}
 	}