Fixes the regression of GCEPD not provisioning correctly on GKE alpha clusters.

[verult]

Fixes the regression by better distinguishing between single-zone and multi-zone PDs.

Which issue(s) this PR fixes (optional, in fixes #<issue number>(, fixes #<issue_number>, ...) format, will close the issue(s) when PR gets merged):
Fixes #59047

Special notes for your reviewer: All PD and dynamic provisioning e2e tests pass, manually verified provision, delete, attach, and detach of both single-zone and multi-zone PDs. Will create e2e tests for multizone PDs in a separate PR.

Release note:

Bug fix: Clusters with GCE feature 'DiskAlphaAPI' enabled were unable to dynamically provision GCE PD volumes.

/sig storage
/assign @saad-ali

[verult]

/release-note-none

[saad-ali]

Since this method is used by both CreateDiskOnCloudProvider(...) and CreateRegionalDiskOnCloudProvider(...), we should probably not key off AlphaFeatureGate.Enabled(AlphaFeatureGCEDisk) alone, right? That would force standard disks that are running on alpha cluster to use the alpha api?

[verult]

Agreed, better to pass that info in from the CreateDisk... methods.

[verult]

Addressed comments, and added basic attach unit test.

[verult]

Test issue: #59518

[saad-ali]

/lgtm
/approve

Could you manually test and verify that RePD is working (create, attach, mount, etc.). And also verify that in a project that is not whitelisted for GCE alpha that standard PD works (create, attach, mount), but RePD fails?

[fejta-bot]

/retest
This bot automatically retries jobs that failed/flaked on approved PRs (send feedback to fejta).

Review the full test history for this PR.

Silence the bot with an /lgtm cancel comment for consistent failures.

[cblecker]

Legitimate verify-bazel failure:

I0208 05:12:20.879] Verifying verify-bazel.sh
I0208 05:12:20.894] 
I0208 05:12:20.896] +++ Running case: verify.bazel 
I0208 05:12:20.897] +++ working dir: /go/src/k8s.io/kubernetes
I0208 05:12:20.899] +++ command: bash "hack/make-rules/../../hack/verify-bazel.sh"
I0208 05:12:39.293] diff -Naupr /go/src/k8s.io/kubernetes/pkg/volume/gce_pd/BUILD /tmp/verify-bazel.fFjFv4/go/src/k8s.io/kubernetes/pkg/volume/gce_pd/BUILD
I0208 05:12:39.293] --- /go/src/k8s.io/kubernetes/pkg/volume/gce_pd/BUILD	2018-02-08 05:08:43.249678444 +0000
I0208 05:12:39.294] +++ /tmp/verify-bazel.fFjFv4/go/src/k8s.io/kubernetes/pkg/volume/gce_pd/BUILD	2018-02-08 05:12:33.130579654 +0000
I0208 05:12:39.294] @@ -43,6 +43,7 @@ go_test(
I0208 05:12:39.295]      embed = [":go_default_library"],
I0208 05:12:39.295]      importpath = "k8s.io/kubernetes/pkg/volume/gce_pd",
I0208 05:12:39.296]      deps = [
I0208 05:12:39.296] +        "//pkg/kubelet/apis:go_default_library",
I0208 05:12:39.297]          "//pkg/util/mount:go_default_library",
I0208 05:12:39.297]          "//pkg/volume:go_default_library",
I0208 05:12:39.297]          "//pkg/volume/testing:go_default_library",
I0208 05:12:39.298] 
I0208 05:12:39.298] Run ./hack/update-bazel.sh
I0208 05:12:40.146] +++ exit code: 1
I0208 05:12:40.150] +++ error: 1
I0208 05:12:40.170] FAILED   verify-bazel.sh	20s

[fejta-bot]

/retest
This bot automatically retries jobs that failed/flaked on approved PRs (send feedback to fejta).

Review the full test history for this PR.

Silence the bot with an /lgtm cancel comment for consistent failures.

[fejta-bot]

/retest
This bot automatically retries jobs that failed/flaked on approved PRs (send feedback to fejta).

Review the full test history for this PR.

Silence the bot with an /lgtm cancel comment for consistent failures.

[cblecker]

/lgtm cancel
Removing LGTM to stop retesting loop

[verult]

@saad-ali I've done the following tests manually for multizone clusters, with DiskAlphaAPI both set and unset, and single-zone cluster with alpha flag unset:

• (Create) Dynamic provisioning with RePD storage class and Delete reclaim policy.
• (Attach & mount) Running a pod referencing the PVC created above.
• (Detach & unmount) Deleting the previous pod
• (Delete) Deleting the PVC created above.
  And all of the above are done using the default gce-pd storage class as well.

[saad-ali]

Awesome, one more thing to run the same set of tests against an alpha cluster on a project that is not whitelisted for the GCE API (to verify the regression is fixed).

/lgtm

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: saad-ali, verult

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these OWNERS Files:

• pkg/cloudprovider/providers/gce/OWNERS [saad-ali]
• pkg/volume/gce_pd/OWNERS [saad-ali]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[saad-ali]

Once merged, this should be cherry picked to 1.9 and 1.8 branches.

[cblecker]

/kind bug
/priority important-soon

[k8s-github-robot]

[MILESTONENOTIFIER] Milestone Pull Request Needs Approval

@cblecker @saad-ali @verult @kubernetes/sig-storage-misc

Action required: This pull request must have the status/approved-for-milestone label applied by a SIG maintainer. If the label is not applied within 6 days, the pull request will be moved out of the v1.9 milestone.

Pull Request Labels

• sig/storage: Pull Request will be escalated to these SIGs if needed.
• priority/important-soon: Escalate to the pull request owners and SIG owner; move out of milestone after several unsuccessful escalation attempts.
• kind/bug: Fixes a bug discovered during the current release.

Help

• Additional instructions
• Commands for setting labels

[k8s-github-robot]

/test all [submit-queue is verifying that this PR is safe to merge]

[k8s-github-robot]

Automatic merge from submit-queue. If you want to cherry-pick this change to another branch, please follow the instructions here.

[k8s-cherrypick-bot]

Commit found in the "release-1.9" branch appears to be this PR. Removing the "cherrypick-candidate" label. If this is an error find help to get your PR picked.