diff --git a/test/e2e/framework.go b/test/e2e/framework.go
index b2167d7f1bb..de9b69406a5 100644
--- a/test/e2e/framework.go
+++ b/test/e2e/framework.go
@@ -94,7 +94,8 @@ func (f *Framework) afterEach() {
 	}
 
 	By(fmt.Sprintf("Destroying namespace %q for this suite.", f.Namespace.Name))
-	if err := f.Client.Namespaces().Delete(f.Namespace.Name); err != nil {
+
+	if err := deleteNS(f.Client, f.Namespace.Name); err != nil {
 		Failf("Couldn't delete ns %q: %s", f.Namespace.Name, err)
 	}
 	// Paranoia-- prevent reuse!
diff --git a/test/e2e/kubectl.go b/test/e2e/kubectl.go
index a7ca944f2d7..02d289ba257 100644
--- a/test/e2e/kubectl.go
+++ b/test/e2e/kubectl.go
@@ -78,7 +78,7 @@ var _ = Describe("Kubectl client", func() {
 
 	AfterEach(func() {
 		By(fmt.Sprintf("Destroying namespace for this suite %v", ns))
-		if err := c.Namespaces().Delete(ns); err != nil {
+		if err := deleteNS(c, ns); err != nil {
 			Failf("Couldn't delete ns %s", err)
 		}
 	})
@@ -470,7 +470,7 @@ var _ = Describe("Kubectl client", func() {
 					}
 				}
 				if !found {
-					Failf("Added annation not found")
+					Failf("Added annotation not found")
 				}
 			})
 		})
diff --git a/test/e2e/pods.go b/test/e2e/pods.go
index bf28677edb8..4fc7b2d70d8 100644
--- a/test/e2e/pods.go
+++ b/test/e2e/pods.go
@@ -246,14 +246,11 @@ var _ = Describe("Pods", func() {
 		}
 
 		By("deleting the pod")
-		podClient.Delete(pod.Name, nil)
-		pods, err = podClient.List(labels.SelectorFromSet(labels.Set(map[string]string{"time": value})), fields.Everything())
-		if err != nil {
+		if err := podClient.Delete(pod.Name, nil); err != nil {
 			Failf("Failed to delete pod: %v", err)
 		}
-		Expect(len(pods.Items)).To(Equal(0))
 
-		By("veryfying pod deletion was observed")
+		By("verifying pod deletion was observed")
 		deleted := false
 		timeout := false
 		timer := time.After(podStartTimeout)
@@ -270,6 +267,12 @@ var _ = Describe("Pods", func() {
 		if !deleted {
 			Fail("Failed to observe pod deletion")
 		}
+
+		pods, err = podClient.List(labels.SelectorFromSet(labels.Set(map[string]string{"time": value})), fields.Everything())
+		if err != nil {
+			Fail(fmt.Sprintf("Failed to list pods to verify deletion: %v", err))
+		}
+		Expect(len(pods.Items)).To(Equal(0))
 	})
 
 	It("should be updated", func() {
diff --git a/test/e2e/service.go b/test/e2e/service.go
index 5a18e576bea..0ffaaf29fa1 100644
--- a/test/e2e/service.go
+++ b/test/e2e/service.go
@@ -64,7 +64,7 @@ var _ = Describe("Services", func() {
 	AfterEach(func() {
 		for _, ns := range namespaces {
 			By(fmt.Sprintf("Destroying namespace %v", ns))
-			if err := c.Namespaces().Delete(ns); err != nil {
+			if err := deleteNS(c, ns); err != nil {
 				Failf("Couldn't delete namespace %s: %s", ns, err)
 			}
 		}
@@ -1121,6 +1121,14 @@ func validateEndpointsOrFail(c *client.Client, namespace, serviceName string, ex
 
 		Logf("Unexpected number of endpoints: found %v, expected %v (%v elapsed, ignoring for 5s)", portsByPodUID, expectedEndpoints, time.Since(start))
 	}
+
+	if pods, err := c.Pods(api.NamespaceAll).List(labels.Everything(), fields.Everything()); err == nil {
+		for _, pod := range pods.Items {
+			Logf("Pod %s\t%s\t%s\t%s", pod.Namespace, pod.Name, pod.Spec.NodeName, pod.DeletionTimestamp)
+		}
+	} else {
+		Logf("Can't list pod debug info: %v", err)
+	}
 	Failf("Timed out waiting for service %s in namespace %s to expose endpoints %v (%v elapsed)", serviceName, namespace, expectedEndpoints, serviceStartTimeout)
 }
 
diff --git a/test/e2e/util.go b/test/e2e/util.go
index 51ecabf891b..0b6ff318504 100644
--- a/test/e2e/util.go
+++ b/test/e2e/util.go
@@ -501,7 +501,7 @@ func deleteTestingNS(c *client.Client) error {
 		for _, ns := range namespaces.Items {
 			if strings.HasPrefix(ns.ObjectMeta.Name, "e2e-tests-") {
 				if ns.Status.Phase == api.NamespaceActive {
-					return fmt.Errorf("Namespace %s is active", ns)
+					return fmt.Errorf("Namespace %s is active", ns.ObjectMeta.Name)
 				}
 				terminating++
 			}
@@ -513,6 +513,51 @@ func deleteTestingNS(c *client.Client) error {
 	return fmt.Errorf("Waiting for terminating namespaces to be deleted timed out")
 }
 
+// deleteNS deletes the provided namespace, waits for it to be completely deleted, and then checks
+// whether there are any pods remaining in a non-terminating state.
+func deleteNS(c *client.Client, namespace string) error {
+	if err := c.Namespaces().Delete(namespace); err != nil {
+		return err
+	}
+
+	err := wait.Poll(1*time.Second, 2*time.Minute, func() (bool, error) {
+		if _, err := c.Namespaces().Get(namespace); err != nil {
+			if apierrs.IsNotFound(err) {
+				return true, nil
+			}
+			Logf("Error while waiting for namespace to be terminated: %v", err)
+			return false, nil
+		}
+		return false, nil
+	})
+
+	// check for pods that were not deleted
+	remaining := []string{}
+	missingTimestamp := false
+	if pods, perr := c.Pods(namespace).List(labels.Everything(), fields.Everything()); perr == nil {
+		for _, pod := range pods.Items {
+			Logf("Pod %s %s on node %s remains, has deletion timestamp %s", namespace, pod.Name, pod.Spec.NodeName, pod.DeletionTimestamp)
+			remaining = append(remaining, pod.Name)
+			if pod.DeletionTimestamp == nil {
+				missingTimestamp = true
+			}
+		}
+	}
+
+	// a timeout occured
+	if err != nil {
+		if missingTimestamp {
+			return fmt.Errorf("namespace %s was not deleted within limit: %v, some pods were not marked with a deletion timestamp, pods remaining: %v", namespace, err, remaining)
+		}
+		return fmt.Errorf("namespace %s was not deleted within limit: %v, pods remaining: %v", namespace, err, remaining)
+	}
+	// pods were not deleted but the namespace was deleted
+	if len(remaining) > 0 {
+		return fmt.Errorf("pods remained within namespace %s after deletion: %v", namespace, remaining)
+	}
+	return nil
+}
+
 func waitForPodRunningInNamespace(c *client.Client, podName string, namespace string) error {
 	return waitForPodCondition(c, namespace, podName, "running", podStartTimeout, func(pod *api.Pod) (bool, error) {
 		if pod.Status.Phase == api.PodRunning {
@@ -549,12 +594,10 @@ func waitForPodSuccessInNamespace(c *client.Client, podName string, contName str
 				if ci.State.Terminated.ExitCode == 0 {
 					By("Saw pod success")
 					return true, nil
-				} else {
-					return true, fmt.Errorf("pod '%s' terminated with failure: %+v", podName, ci.State.Terminated)
 				}
-			} else {
-				Logf("Nil State.Terminated for container '%s' in pod '%s' in namespace '%s' so far", contName, podName, namespace)
+				return true, fmt.Errorf("pod '%s' terminated with failure: %+v", podName, ci.State.Terminated)
 			}
+			Logf("Nil State.Terminated for container '%s' in pod '%s' in namespace '%s' so far", contName, podName, namespace)
 		}
 		return false, nil
 	})
@@ -963,7 +1006,7 @@ func tryKill(cmd *exec.Cmd) {
 func testContainerOutputInNamespace(scenarioName string, c *client.Client, pod *api.Pod, containerIndex int, expectedOutput []string, ns string) {
 	By(fmt.Sprintf("Creating a pod to test %v", scenarioName))
 
-	defer c.Pods(ns).Delete(pod.Name, nil)
+	defer c.Pods(ns).Delete(pod.Name, api.NewDeleteOptions(0))
 	if _, err := c.Pods(ns).Create(pod); err != nil {
 		Failf("Failed to create pod: %v", err)
 	}
@@ -1247,6 +1290,13 @@ func RunRC(config RCConfig) error {
 	}
 
 	if oldRunning != config.Replicas {
+		if pods, err := config.Client.Pods(api.NamespaceAll).List(labels.Everything(), fields.Everything()); err == nil {
+			for _, pod := range pods.Items {
+				Logf("Pod %s\t%s\t%s\t%s", pod.Namespace, pod.Name, pod.Spec.NodeName, pod.DeletionTimestamp)
+			}
+		} else {
+			Logf("Can't list pod debug info: %v", err)
+		}
 		return fmt.Errorf("Only %d pods started out of %d", oldRunning, config.Replicas)
 	}
 	return nil