TestFunctional/parallel/ComponentHealth flake on docker

[priyawadhwa]

Original logs here

functional_test.go:288: (dbg) Run:  kubectl --context functional-20200505192422-29695 get cs -o=json
functional_test.go:307: unexpected status: False - item: {TypeMeta:{Kind:ComponentStatus APIVersion:v1} ObjectMeta:{Name:controller-manager GenerateName: Namespace: SelfLink:/api/v1/componentstatuses/controller-manager UID: ResourceVersion: Generation:0 CreationTimestamp:0001-01-01 00:00:00 +0000 UTC DeletionTimestamp:<nil> DeletionGracePeriodSeconds:<nil> Labels:map[] Annotations:map[]} Conditions:[{Type:Healthy Status:False Message:Get http://127.0.0.1:10252/healthz: dial tcp 127.0.0.1:10252: connect: connection refused Error:}]}

[medyagh]

I still see this on HEAD today

[medyagh]

seen again today, with more logging https://storage.googleapis.com/minikube-builds/logs/7959/3fa4b91/Docker_Linux.html

[priyawadhwa]

According to Thomas's passrate script, this test has been a little flaky but passed on docker today:

$ /home/tstromberg/bin/passrate.sh TestFunctional/parallel/ComponentHealth
-------------------------------------------------
docker_Linux      | 2020-05-14 |  67% | 8 of 12
docker_Linux      | 2020-05-15 |  75% | 9 of 12
docker_Linux      | 2020-05-16 |  75% | 9 of 12
docker_Linux      | 2020-05-17 |  92% | 11 of 12
docker_Linux      | 2020-05-18 | 100% | 8 of 8

it also seemed to fail once today on KVM

KVM_Linux         | 2020-05-14 | 100% | 10 of 10
KVM_Linux         | 2020-05-15 |  92% | 11 of 12
KVM_Linux         | 2020-05-16 |  92% | 12 of 13
KVM_Linux         | 2020-05-17 | 100% | 11 of 11
KVM_Linux         | 2020-05-18 |  88% | 7 of 8

I'll do some digging.

To run locally:

make integration -e TEST_ARGS="-test.run TestFunctional/parallel/ComponentHealth --profile=minikube --cleanup=false --minikube-start-args="--driver=docker""

[priyawadhwa]

Most recent logs from #8168 can be found here