Wait for pod running status before timeout despite errors of accessing API server

Signed-off-by: danfengl <danfengl@vmware.com>

Dockerfile

@@ -11,7 +11,7 @@
 # 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.
-FROM --platform=$BUILDPLATFORM golang:1.17.11 as builder-env
+FROM --platform=$BUILDPLATFORM golang:1.17 as builder-env
 
 ARG GOPROXY
 ARG PKG
@@ -50,7 +50,7 @@ RUN mkdir -p /output/usr/bin && \
     go build -o /output/${BIN} \
     -ldflags "${LDFLAGS}" ${PKG}/cmd/${BIN}
 
-FROM gcr.io/distroless/base-debian11@sha256:e672eb713e56feb13e349773973b81b1b9284f70b15cf18d1a09ad31a03abe59
+FROM gcr.io/distroless/base-debian11:nonroot
 
 LABEL maintainer="Nolan Brubaker <brubakern@vmware.com>"
 

hack/build-image/Dockerfile

@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-FROM golang:1.17.11
+FROM golang:1.17
 
 ARG GOPROXY
 

test/e2e/util/k8s/common.go

@@ -62,7 +62,9 @@ func WaitForPods(ctx context.Context, client TestClient, namespace string, pods
 		for _, podName := range pods {
 			checkPod, err := client.ClientGo.CoreV1().Pods(namespace).Get(ctx, podName, metav1.GetOptions{})
 			if err != nil {
-				return false, errors.WithMessage(err, fmt.Sprintf("Failed to verify pod %s/%s is %s", namespace, podName, corev1api.PodRunning))
+				//Should ignore "etcdserver: request timed out" kind of errors, try to get pod status again before timeout.
+				fmt.Println(errors.Wrap(err, fmt.Sprintf("Failed to verify pod %s/%s is %s, try again...", namespace, podName, corev1api.PodRunning)))
+				return false, nil
 			}
 			// If any pod is still waiting we don't need to check any more so return and wait for next poll interval
 			if checkPod.Status.Phase != corev1api.PodRunning {