helm: Use batch/v1 apiVersion for CronJob in K8s 1.21+

[gandro]

The CronJob object reached GA in Kubernetes 1.2, thus bumping its api
version from batch/v1beta1 to batch/v1. The old batch/v1beta1 api
version will be removed in Kubernetes 1.25.

This fixes the following Helm warning on Kubernetes 1.21+:

W0623 11:06:59.097404 1094484 warnings.go:67] batch/v1beta1 CronJob is deprecated in v1.21+, unavailable in v1.25+; use batch/v1 CronJob

The first commit is a drive-by fixup for 4cc699d (see commit description).

[gandro]

test-me-please

[gandro]

Tests failed due the branch being too old (rebased CI tests are using flags that don't exist in my branch). Rebasing and restarting Jenkins tests.

[gandro]

test-me-please

[gandro]

Jenkins is green:

Will update the commit message to address the typo without re-running Jenkins. Other than that, it is ready to merge as all required tests have passed.

[pchaigno]

This pull request doesn't seem to be among the exceptions for the merge freeze, so removing ready-to-merge label.

[gandro]

This pull request doesn't seem to be among the exceptions for the merge freeze, so removing ready-to-merge label.

This is because of Travis failing, right?

Edit: Ah, I guess it's because of master having flakes. But I guess the Travis flake I hit should also be fixed before we merge this.

[gandro]

This needs a rebase, but I'm holding off for #16900 to avoid further conflicts.

Also removing the merge freeze label, as the occuring Travis flake got fixed and my understanding is that master is not frozen anymore.

[stale]

This pull request has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[pchaigno]

This needs a rebase, but I'm holding off for #16900 to avoid further conflicts.

#16900 was merged 🎉

[gandro]

test-me-please

Job 'Cilium-PR-K8s-1.20-kernel-4.19' failed and has not been observed before, so may be related to your PR:

Click to show.

Test Name

K8sServicesTest Checks ClusterIP Connectivity Checks service.kubernetes.io/service-proxy-name label implementation

Failure Output

FAIL: Request from echo-8fd54d9fd-4f8pz pod to service http://10.100.162.212:80/ failed

If it is a flake, comment /mlh new-flake Cilium-PR-K8s-1.20-kernel-4.19 so I can create a new GitHub issue to track it.

Job 'Cilium-PR-K8s-GKE' failed and has not been observed before, so may be related to your PR:

Click to show.

Test Name

K8sDatapathConfig MonitorAggregation Checks that monitor aggregation flags send notifications

Failure Output

FAIL: Pods are not ready in time: timed out waiting for pods with filter  to be ready: 4m0s timeout expired

If it is a flake, comment /mlh new-flake Cilium-PR-K8s-GKE so I can create a new GitHub issue to track it.

[gandro]

/mlh new-flake Cilium-PR-K8s-1.20-kernel-4.19

👍 created #17403

---

Edit the other ones look like flakes as well:

• GKE (https://jenkins.cilium.io/job/Cilium-PR-K8s-GKE/6392/) seems to have hit CI: gke-stable (test-gke): Pods are not ready in time: connection refused #17307
• 1.21-kernel-4.9 (https://jenkins.cilium.io/job/Cilium-PR-K8s-1.16-net-next/1429/) hit CI: K8sPolicyTest Basic Test checks policies with named ports #17232
• 1.16-netnext (https://jenkins.cilium.io/job/Cilium-PR-K8s-1.16-net-next/1429/) hit CI: K8sServicesTest Checks fails with "Can not connect to service X from outside cluster" #12511 and some new policy flakes (CI: K8sPolicyTest Basic Test Traffic redirections to proxy Tests #17404)

[gandro]

I believe the tests in question have since been disabled. Rebasing and re-running tests.

[gandro]

/test

• Runtime hit a provisioning failure https://jenkins.cilium.io/job/Cilium-PR-Runtime-net-next/362/
• AKS (https://github.com/cilium/cilium/actions/runs/1353833219) hit an issue with to-fqdns tests
• GKE (https://github.com/cilium/cilium/runs/3924427118?check_suite_focus=true) hit flake: timeout reached waiting for service (echo-same-node or echo-other-node) cilium-cli#342
• Multicluster (https://github.com/cilium/cilium/runs/3924425732?check_suite_focus=true) hit a connection timeout while accessing echo-other-node

[gandro]

I'm marking this ready to merge. It has been open for 4 months now (blocked due to the zero-flake strategy) and is a rather trivial change in the Helm chart. It has been blocked by various CI flakes, many of which have been fixed or disabled, but new ones pop up each new run.

Since all recently failed test suites in the last test run do not contain any Helm-related tests, I am fairly confident that those are also unrelated flakes and that this PR does not introduce any regressions to CI.