distributed-on-master failing

[chrishenzie]

The distributed-on-master test has been failing since April 7th.

Taking a sample error, it looks like the root cause is a missing test-driver.yaml:

sed: can't read /home/prow/go/pkg/csiprow.z0lNIIs8xX/test-driver.yaml: No such file or directory

cc @msau42 @pohly

[pohly]

I have a fix for that in PR #253

I know, should have been a separate PR, but I was expecting that one to merge a bit sooner.

[chrishenzie]

@pohly Thanks for the speedy fix! It looks like there was one more test failure in the most recent run.

[pohly]

It looks like there was one more test failure in the most recent run.

That looks like a timing flake. Let's keep this issue open and monitor that a bit longer.

/reopen

[k8s-ci-robot]

@pohly: Reopened this issue.

In response to this:

It looks like there was one more test failure in the most recent run.

That looks like a timing flake. Let's keep this issue open and monitor that a bit longer.

/reopen

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[pohly]

Hmm, occurred twice already. I'll check...

[pohly]

External Storage [Driver: hostpath.csi.k8s.io] [Testpattern: Dynamic PV (immediate binding)] topology should provision a volume and schedule a pod with AllowedTopologies is still flaky.

Let's investigate as part of this issue.

[pohly]

The reason for that failure was identified as problem in external-provisioner and fixed in kubernetes-csi/external-provisioner#612.

We can fix the job by updating to the latest external-provisioner.

[pohly]

After merging PR #280, the job started to pass: https://prow.k8s.io/view/gs/kubernetes-jenkins/logs/ci-kubernetes-csi-distributed-on-kubernetes-master/1389537313842794496