diff --git a/pkg/kubelet/kubelet.go b/pkg/kubelet/kubelet.go
index 59d4158ee8f..a5474f8d702 100644
--- a/pkg/kubelet/kubelet.go
+++ b/pkg/kubelet/kubelet.go
@@ -1883,7 +1883,7 @@ func (kl *Kubelet) SyncPod(ctx context.Context, updateType kubetypes.SyncPodType
 	// handlePodResourcesResize updates the pod to use the allocated resources. This should come
 	// before the main business logic of SyncPod, so that a consistent view of the pod is used
 	// across the sync loop.
-	if kuberuntime.IsInPlacePodVerticalScalingAllowed(pod) {
+	if utilfeature.DefaultFeatureGate.Enabled(features.InPlacePodVerticalScaling) {
 		// Handle pod resize here instead of doing it in HandlePodUpdates because
 		// this conveniently retries any Deferred resize requests
 		// TODO(vinaykul,InPlacePodVerticalScaling): Investigate doing this in HandlePodUpdates + periodic SyncLoop scan
@@ -2816,10 +2816,6 @@ func (kl *Kubelet) HandlePodSyncs(pods []*v1.Pod) {
 // pod should hold the desired (pre-allocated) spec.
 // Returns true if the resize can proceed.
 func (kl *Kubelet) canResizePod(pod *v1.Pod) (bool, v1.PodResizeStatus, string) {
-	if goos == "windows" {
-		return false, v1.PodResizeStatusInfeasible, "Resizing Windows pods is not supported"
-	}
-
 	if v1qos.GetPodQOS(pod) == v1.PodQOSGuaranteed && !utilfeature.DefaultFeatureGate.Enabled(features.InPlacePodVerticalScalingExclusiveCPUs) {
 		if utilfeature.DefaultFeatureGate.Enabled(features.CPUManager) {
 			if kl.containerManager.GetNodeConfig().CPUManagerPolicy == "static" {
@@ -2877,6 +2873,27 @@ func (kl *Kubelet) canResizePod(pod *v1.Pod) (bool, v1.PodResizeStatus, string)
 // the allocation decision and pod status.
 func (kl *Kubelet) handlePodResourcesResize(pod *v1.Pod, podStatus *kubecontainer.PodStatus) (*v1.Pod, error) {
 	allocatedPod, updated := kl.allocationManager.UpdatePodFromAllocation(pod)
+	// Keep this logic in sync with kuberuntime.isInPlacePodVerticalScalingAllowed
+	if goos == "windows" || kubetypes.IsStaticPod(pod) {
+		if updated {
+			// A resize is requested but not supported.
+			var msg string
+			switch {
+			case goos == "windows":
+				msg = "Resizing Windows pods is not supported"
+			case kubetypes.IsStaticPod(pod):
+				msg = "Resizing static pods is not supported"
+			default:
+				msg = "Resizing this pod is not supported"
+			}
+			kl.recorder.Eventf(pod, v1.EventTypeWarning, events.ResizeInfeasible, msg)
+			kl.statusManager.SetPodResizeStatus(pod.UID, v1.PodResizeStatusInfeasible)
+		} else {
+			kl.statusManager.SetPodResizeStatus(pod.UID, "")
+		}
+		return allocatedPod, nil
+	}
+
 	if !updated {
 		// Desired resources == allocated resources. Check whether a resize is in progress.
 		resizeInProgress := !allocatedResourcesMatchStatus(allocatedPod, podStatus)
diff --git a/pkg/kubelet/kubelet_pods.go b/pkg/kubelet/kubelet_pods.go
index 28378727e87..c26138a11cb 100644
--- a/pkg/kubelet/kubelet_pods.go
+++ b/pkg/kubelet/kubelet_pods.go
@@ -1748,10 +1748,6 @@ func getPhase(pod *v1.Pod, info []v1.ContainerStatus, podIsTerminal bool) v1.Pod
 }
 
 func (kl *Kubelet) determinePodResizeStatus(allocatedPod *v1.Pod, podStatus *kubecontainer.PodStatus, podIsTerminal bool) v1.PodResizeStatus {
-	if kubetypes.IsStaticPod(allocatedPod) {
-		return ""
-	}
-
 	// If pod is terminal, clear the resize status.
 	if podIsTerminal {
 		kl.statusManager.SetPodResizeStatus(allocatedPod.UID, "")
diff --git a/pkg/kubelet/kubelet_test.go b/pkg/kubelet/kubelet_test.go
index f46defb0d03..f3c617a9a60 100644
--- a/pkg/kubelet/kubelet_test.go
+++ b/pkg/kubelet/kubelet_test.go
@@ -2712,6 +2712,7 @@ func TestHandlePodResourcesResize(t *testing.T) {
 		expectedResize        v1.PodResizeStatus
 		expectBackoffReset    bool
 		goos                  string
+		annotations           map[string]string
 	}{
 		{
 			name:                  "Request CPU and memory decrease - expect InProgress",
@@ -2788,6 +2789,14 @@ func TestHandlePodResourcesResize(t *testing.T) {
 			expectedResize:        v1.PodResizeStatusInfeasible,
 			goos:                  "windows",
 		},
+		{
+			name:                  "static pod, expect Infeasible",
+			originalRequests:      v1.ResourceList{v1.ResourceCPU: cpu1000m, v1.ResourceMemory: mem1000M},
+			newRequests:           v1.ResourceList{v1.ResourceCPU: cpu500m, v1.ResourceMemory: mem500M},
+			expectedAllocatedReqs: v1.ResourceList{v1.ResourceCPU: cpu1000m, v1.ResourceMemory: mem1000M},
+			expectedResize:        v1.PodResizeStatusInfeasible,
+			annotations:           map[string]string{kubetypes.ConfigSourceAnnotationKey: kubetypes.FileSource},
+		},
 		{
 			name:                  "Increase CPU from min shares",
 			originalRequests:      v1.ResourceList{v1.ResourceCPU: cpu2m},
@@ -2889,6 +2898,7 @@ func TestHandlePodResourcesResize(t *testing.T) {
 					originalPod = testPod1.DeepCopy()
 					originalCtr = &originalPod.Spec.Containers[0]
 				}
+				originalPod.Annotations = tt.annotations
 				originalCtr.Resources.Requests = tt.originalRequests
 				originalCtr.Resources.Limits = tt.originalLimits
 
diff --git a/pkg/kubelet/kuberuntime/kuberuntime_manager.go b/pkg/kubelet/kuberuntime/kuberuntime_manager.go
index 0074688aa6a..03c4ad8d866 100644
--- a/pkg/kubelet/kuberuntime/kuberuntime_manager.go
+++ b/pkg/kubelet/kuberuntime/kuberuntime_manager.go
@@ -551,7 +551,7 @@ func containerSucceeded(c *v1.Container, podStatus *kubecontainer.PodStatus) boo
 	return cStatus.State == kubecontainer.ContainerStateExited && cStatus.ExitCode == 0
 }
 
-func IsInPlacePodVerticalScalingAllowed(pod *v1.Pod) bool {
+func isInPlacePodVerticalScalingAllowed(pod *v1.Pod) bool {
 	return utilfeature.DefaultFeatureGate.Enabled(features.InPlacePodVerticalScaling) &&
 		!types.IsStaticPod(pod) &&
 		runtime.GOOS != "windows"
@@ -561,7 +561,7 @@ func IsInPlacePodVerticalScalingAllowed(pod *v1.Pod) bool {
 // Returns whether to keep (true) or restart (false) the container.
 // TODO(vibansal): Make this function to be agnostic to whether it is dealing with a restartable init container or not (i.e. remove the argument `isRestartableInitContainer`).
 func (m *kubeGenericRuntimeManager) computePodResizeAction(pod *v1.Pod, containerIdx int, isRestartableInitContainer bool, kubeContainerStatus *kubecontainer.Status, changes *podActions) (keepContainer bool) {
-	if !IsInPlacePodVerticalScalingAllowed(pod) {
+	if !isInPlacePodVerticalScalingAllowed(pod) {
 		return true
 	}
 
@@ -998,7 +998,7 @@ func (m *kubeGenericRuntimeManager) computePodActions(ctx context.Context, pod *
 		}
 	}
 
-	if IsInPlacePodVerticalScalingAllowed(pod) {
+	if isInPlacePodVerticalScalingAllowed(pod) {
 		changes.ContainersToUpdate = make(map[v1.ResourceName][]containerToUpdateInfo)
 	}
 
@@ -1414,7 +1414,7 @@ func (m *kubeGenericRuntimeManager) SyncPod(ctx context.Context, pod *v1.Pod, po
 	}
 
 	// Step 7: For containers in podContainerChanges.ContainersToUpdate[CPU,Memory] list, invoke UpdateContainerResources
-	if IsInPlacePodVerticalScalingAllowed(pod) {
+	if isInPlacePodVerticalScalingAllowed(pod) {
 		if len(podContainerChanges.ContainersToUpdate) > 0 || podContainerChanges.UpdatePodResources {
 			m.doPodResizeAction(pod, podContainerChanges, &result)
 		}