diff --git a/test/e2e/resize_nodes.go b/test/e2e/resize_nodes.go
index d0bfe9bfadd..25383a07821 100644
--- a/test/e2e/resize_nodes.go
+++ b/test/e2e/resize_nodes.go
@@ -185,6 +185,7 @@ func rcByNamePort(name string, replicas int, image string, port int, labels map[
 func rcByNameContainer(name string, replicas int, image string, labels map[string]string, c api.Container) *api.ReplicationController {
 	// Add "name": name to the labels, overwriting if it exists.
 	labels["name"] = name
+	gracePeriod := int64(0)
 	return &api.ReplicationController{
 		TypeMeta: unversioned.TypeMeta{
 			Kind:       "ReplicationController",
@@ -203,7 +204,8 @@ func rcByNameContainer(name string, replicas int, image string, labels map[strin
 					Labels: labels,
 				},
 				Spec: api.PodSpec{
-					Containers: []api.Container{c},
+					Containers:                    []api.Container{c},
+					TerminationGracePeriodSeconds: &gracePeriod,
 				},
 			},
 		},
diff --git a/test/e2e/util.go b/test/e2e/util.go
index 49a9696426a..ecebb6463c7 100644
--- a/test/e2e/util.go
+++ b/test/e2e/util.go
@@ -768,7 +768,10 @@ func waitForPodToDisappear(c *client.Client, ns, podName string, label labels.Se
 // In case of failure or too long waiting time, an error is returned.
 func waitForRCPodToDisappear(c *client.Client, ns, rcName, podName string) error {
 	label := labels.SelectorFromSet(labels.Set(map[string]string{"name": rcName}))
-	return waitForPodToDisappear(c, ns, podName, label, 20*time.Second, 5*time.Minute)
+	// NodeController evicts pod after 5 minutes, so we need timeout greater than that.
+	// Additionally, there can be non-zero grace period, so we are setting 10 minutes
+	// to be on the safe size.
+	return waitForPodToDisappear(c, ns, podName, label, 20*time.Second, 10*time.Minute)
 }
 
 // waitForService waits until the service appears (exist == true), or disappears (exist == false)