diff --git a/test/e2e/cluster_upgrade.go b/test/e2e/cluster_upgrade.go
index 4ab6f096da5..6b419867954 100644
--- a/test/e2e/cluster_upgrade.go
+++ b/test/e2e/cluster_upgrade.go
@@ -38,6 +38,7 @@ var upgradeTests = []upgrades.Test{
 	&upgrades.ConfigMapUpgradeTest{},
 	&upgrades.HPAUpgradeTest{},
 	&upgrades.PersistentVolumeUpgradeTest{},
+	&upgrades.DaemonSetUpgradeTest{},
 }
 
 var _ = framework.KubeDescribe("Upgrade [Feature:Upgrade]", func() {
diff --git a/test/e2e/upgrades/BUILD b/test/e2e/upgrades/BUILD
index 7ce61c9002f..e939012ede7 100644
--- a/test/e2e/upgrades/BUILD
+++ b/test/e2e/upgrades/BUILD
@@ -11,6 +11,7 @@ go_library(
     name = "go_default_library",
     srcs = [
         "configmaps.go",
+        "daemonsets.go",
         "deployments.go",
         "horizontal_pod_autoscalers.go",
         "job.go",
@@ -27,6 +28,7 @@ go_library(
         "//pkg/apis/apps/v1beta1:go_default_library",
         "//pkg/apis/batch/v1:go_default_library",
         "//pkg/apis/extensions/v1beta1:go_default_library",
+        "//pkg/controller:go_default_library",
         "//pkg/controller/deployment/util:go_default_library",
         "//pkg/kubelet/sysctl:go_default_library",
         "//test/e2e/common:go_default_library",
@@ -35,6 +37,7 @@ go_library(
         "//vendor:github.com/onsi/gomega",
         "//vendor:k8s.io/apimachinery/pkg/api/errors",
         "//vendor:k8s.io/apimachinery/pkg/apis/meta/v1",
+        "//vendor:k8s.io/apimachinery/pkg/labels",
         "//vendor:k8s.io/apimachinery/pkg/util/uuid",
         "//vendor:k8s.io/apimachinery/pkg/util/wait",
     ],
diff --git a/test/e2e/upgrades/daemonsets.go b/test/e2e/upgrades/daemonsets.go
new file mode 100644
index 00000000000..babdf3fadd5
--- /dev/null
+++ b/test/e2e/upgrades/daemonsets.go
@@ -0,0 +1,181 @@
+/*
+Copyright 2017 The Kubernetes Authors.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+*/
+
+package upgrades
+
+import (
+	"k8s.io/apimachinery/pkg/labels"
+	"k8s.io/apimachinery/pkg/util/wait"
+	"k8s.io/kubernetes/pkg/controller"
+
+	"k8s.io/kubernetes/pkg/api/v1"
+
+	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
+	extensions "k8s.io/kubernetes/pkg/apis/extensions/v1beta1"
+	"k8s.io/kubernetes/test/e2e/framework"
+
+	. "github.com/onsi/ginkgo"
+)
+
+// DaemonSetUpgradeTest tests that a DaemonSet is running before and after
+// a cluster upgrade.
+type DaemonSetUpgradeTest struct {
+	daemonSet *extensions.DaemonSet
+}
+
+// Setup creates a DaemonSet and verifies that it's running
+func (t *DaemonSetUpgradeTest) Setup(f *framework.Framework) {
+	namespace := "daemonset-upgrade"
+	daemonSetName := "ds1"
+	labelSet := map[string]string{"ds-name": daemonSetName}
+	image := "gcr.io/google_containers/serve_hostname:v1.4"
+
+	// Grab a unique namespace so we don't collide.
+	ns, err := f.CreateNamespace(namespace, nil)
+	framework.ExpectNoError(err)
+
+	t.daemonSet = &extensions.DaemonSet{
+		ObjectMeta: metav1.ObjectMeta{
+			Namespace: ns.Name,
+			Name:      daemonSetName,
+		},
+		Spec: extensions.DaemonSetSpec{
+			Template: v1.PodTemplateSpec{
+				ObjectMeta: metav1.ObjectMeta{
+					Labels: labelSet,
+				},
+				Spec: v1.PodSpec{
+					Containers: []v1.Container{
+						{
+							Name:  daemonSetName,
+							Image: image,
+							Ports: []v1.ContainerPort{{ContainerPort: 9376}},
+						},
+					},
+				},
+			},
+		},
+	}
+
+	By("Creating a DaemonSet")
+	if t.daemonSet, err = f.ClientSet.Extensions().DaemonSets(ns.Name).Create(t.daemonSet); err != nil {
+		framework.Failf("unable to create test DaemonSet %s: %v", t.daemonSet.Name, err)
+	}
+
+	By("Waiting for DaemonSet pods to become ready")
+	err = wait.Poll(framework.Poll, framework.PodStartTimeout, func() (bool, error) {
+		return checkRunningOnAllNodes(f, t.daemonSet.Namespace, t.daemonSet.Labels)
+	})
+	framework.ExpectNoError(err)
+
+	By("Validating the DaemonSet after creation")
+	t.validateRunningDaemonSet(f)
+}
+
+// Test waits until the upgrade has completed and then verifies that the DaemonSet
+// is still running
+func (t *DaemonSetUpgradeTest) Test(f *framework.Framework, done <-chan struct{}, upgrade UpgradeType) {
+	By("Waiting for upgradet to complete before re-validating DaemonSet")
+	<-done
+
+	By("validating the DaemonSet is still running after upgrade")
+	t.validateRunningDaemonSet(f)
+}
+
+// Teardown cleans up any remaining resources.
+func (t *DaemonSetUpgradeTest) Teardown(f *framework.Framework) {
+	// rely on the namespace deletion to clean up everything
+}
+
+func (t *DaemonSetUpgradeTest) validateRunningDaemonSet(f *framework.Framework) {
+	By("confirming the DaemonSet pods are running on all expected nodes")
+	res, err := checkRunningOnAllNodes(f, t.daemonSet.Namespace, t.daemonSet.Labels)
+	framework.ExpectNoError(err)
+	if !res {
+		framework.Failf("expected DaemonSet pod to be running on all nodes, it was not")
+	}
+
+	// DaemonSet resource itself should be good
+	By("confirming the DaemonSet resource is in a good state")
+	res, err = checkDaemonStatus(f, t.daemonSet.Namespace, t.daemonSet.Name)
+	framework.ExpectNoError(err)
+	if !res {
+		framework.Failf("expected DaemonSet to be in a good state, it was not")
+	}
+}
+
+func checkRunningOnAllNodes(f *framework.Framework, namespace string, selector map[string]string) (bool, error) {
+	nodeList, err := f.ClientSet.Core().Nodes().List(metav1.ListOptions{})
+	if err != nil {
+		return false, err
+	}
+
+	nodeNames := make([]string, 0)
+	for _, node := range nodeList.Items {
+		if len(node.Spec.Taints) == 0 {
+			nodeNames = append(nodeNames, node.Name)
+		} else {
+			framework.Logf("Node %v not expected to have DaemonSet pod, has taints %v", node.Name, node.Spec.Taints)
+		}
+	}
+
+	return checkDaemonPodOnNodes(f, namespace, selector, nodeNames)
+}
+
+func checkDaemonPodOnNodes(f *framework.Framework, namespace string, labelSet map[string]string, nodeNames []string) (bool, error) {
+	selector := labels.Set(labelSet).AsSelector()
+	options := metav1.ListOptions{LabelSelector: selector.String()}
+	podList, err := f.ClientSet.Core().Pods(namespace).List(options)
+	if err != nil {
+		return false, err
+	}
+	pods := podList.Items
+
+	nodesToPodCount := make(map[string]int)
+	for _, pod := range pods {
+		if controller.IsPodActive(&pod) {
+			framework.Logf("Pod name: %v\t Node Name: %v", pod.Name, pod.Spec.NodeName)
+			nodesToPodCount[pod.Spec.NodeName]++
+		}
+	}
+	framework.Logf("nodesToPodCount: %v", nodesToPodCount)
+
+	// Ensure that exactly 1 pod is running on all nodes in nodeNames.
+	for _, nodeName := range nodeNames {
+		if nodesToPodCount[nodeName] != 1 {
+			return false, nil
+		}
+	}
+
+	// Ensure that sizes of the lists are the same. We've verified that every element of nodeNames is in
+	// nodesToPodCount, so verifying the lengths are equal ensures that there aren't pods running on any
+	// other nodes.
+	return len(nodesToPodCount) == len(nodeNames), nil
+}
+
+func checkDaemonStatus(f *framework.Framework, namespace string, dsName string) (bool, error) {
+	ds, err := f.ClientSet.ExtensionsV1beta1().DaemonSets(namespace).Get(dsName, metav1.GetOptions{})
+	if err != nil {
+		return false, err
+	}
+
+	desired, scheduled, ready := ds.Status.DesiredNumberScheduled, ds.Status.CurrentNumberScheduled, ds.Status.NumberReady
+	if desired != scheduled && desired != ready {
+		return false, nil
+	}
+
+	return true, nil
+}