release-1.8 GKE test failure "Services should work after restarting apiserver" #55244
Closed
Description
In the release1.8 blocking test suite. "[sig-network] Services should work after restarting apiserver [Disruptive]" has started failing regularly with the error:
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/network/service.go:434
Nov 4 02:53:35.323: error restarting apiserver: error running gcloud [container clusters --project=k8s-gke-upg-c1-4-g1-6-up-clu --zone=us-central1-b upgrade e2e-856 --master --cluster-version=1.8.3-beta.0.88+dc6c4c5dcebb7e --quiet]; got error exit status 1, stdout "", stderr "Upgrading e2e-856...\n.....done.\nERROR: (gcloud.container.clusters.upgrade) Operation [<Operation\n endTime: u'2017-11-04T02:53:29.555870594Z'\n name: u'operation-1509764009232-95e55dde'\n operationType: OperationTypeValueValuesEnum(UPGRADE_MASTER, 3)\n selfLink: u'https://test-container.sandbox.googleapis.com/v1/projects/433788295746/zones/us-central1-b/operations/operation-1509764009232-95e55dde'\n startTime: u'2017-11-04T02:53:29.232057635Z'\n status: StatusValueValuesEnum(DONE, 3)\n statusMessage: u'Patch failed'\n targetLink: u'https://test-container.sandbox.googleapis.com/v1/projects/433788295746/zones/us-central1-b/clusters/e2e-856'\n zone: u'us-central1-b'>] finished with error: Patch failed\n"
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/network/service.go:414
The first occurance I can find is in build 841. But it has been failing more consistently since around build 856.