Add optional v1alpha4 and self-hosted management cluster tests

[randomvariable]

Signed-off-by: Naadir Jeewa [email protected]

What type of PR is this?

/kind failing-test
/area testing

What this PR does / why we need it:

Which issue(s) this PR fixes (optional, in fixes #<issue number>(, fixes #<issue_number>, ...) format, will close the issue(s) when PR gets merged):
Part of #2788
Part of #2793

Special notes for your reviewer:

Checklist:

• squashed commits
• includes documentation
• adds unit tests
• adds or updates e2e tests

Release note:

NONE

[randomvariable]

/test ?

[k8s-ci-robot]

@randomvariable: The following commands are available to trigger required jobs:

• /test pull-cluster-api-provider-aws-build
• /test pull-cluster-api-provider-aws-test
• /test pull-cluster-api-provider-aws-verify

The following commands are available to trigger optional jobs:

• /test pull-cluster-api-provider-aws-e2e
• /test pull-cluster-api-provider-aws-e2e-blocking
• /test pull-cluster-api-provider-aws-e2e-conformance
• /test pull-cluster-api-provider-aws-e2e-conformance-with-ci-artifacts
• /test pull-cluster-api-provider-aws-e2e-eks

Use /test all to run the following jobs that were automatically triggered:

• pull-cluster-api-provider-aws-build
• pull-cluster-api-provider-aws-test
• pull-cluster-api-provider-aws-verify

In response to this:

/test ?

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.

[randomvariable]

/test pull-cluster-api-provider-aws-e2e

[randomvariable]

/milestone v1.0.0
/triage accepted
/priority critical-urgent

[randomvariable]

/test pull-cluster-api-provider-aws-e2e

[randomvariable]

/test pull-cluster-api-provider-aws-e2e

[randomvariable]

/test pull-cluster-api-provider-aws-e2e

[randomvariable]

/test pull-cluster-api-provider-aws-e2e

[sedefsavas]

I am able to replicate locally. After the upgrade, all deployment machines come up but MachineDeployment is not being reconciled hence never detects scaling up is successful.

[randomvariable]

I am able to replicate locally. After the upgrade, all deployment machines come up but MachineDeployment is not being reconciled hence never detects scaling up is successful.

Is that a separate issue to the one that keeps showing up for all the ones using a remote management cluster?

"reason": "ProgressDeadlineExceeded",
[1] "message": "ReplicaSet "capa-controller-manager-7bd56cd58" has timed out progressing."

[sedefsavas]

They are the same, that deadline is exceeded because CAPI controllers don't reconcile anything.
KCP show only one error:
KubeadmControlPlane" "cluster"="clusterctl-upgrade-vexwe7" "name"="clusterctl-upgrade-vexwe7-control-plane" "namespace"="clusterctl-upgrade" "reconciler group"="controlplane.cluster.x-k8s.io" "reconciler kind"="KubeadmControlPlane" E1008 06:28:50.096627 1 reflector.go:138] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:167: Failed to watch *v1.PartialObjectMetadata: unknown

And CAPI logs do not show any error but also it doesn't reconcile when I make changes to MachineDeployment.

I will check with Stefan/Fabrizio when they are online.

[UPDATE]: Failures I saw was probably due to using an unsupported version of K8s.

[randomvariable]

If you trace the lines back, then it's waiting for the deployment of CAPA controllers to succeed with the appv1.Deployment of the controller coming up. Following the trace to

[1]     /home/prow/go/pkg/mod/sigs.k8s.io/cluster-api/[email protected]/framework/deployment_helpers.go:72
[1] 
[1]     Full Stack Trace
[1]     sigs.k8s.io/cluster-api/test/framework.WaitForDeploymentsAvailable(0x3096840, 0xc000054140, 0x7fe5f0056d40, 0xc0006e40e0, 0xc0003b1400, 0xc001215d80, 0x2, 0x2)
[1]     	/home/prow/go/pkg/mod/sigs.k8s.io/cluster-api/[email protected]/framework/deployment_helpers.go:72 +0x2cf
[1]     sigs.k8s.io/cluster-api/test/framework/clusterctl.UpgradeManagementClusterAndWait(0x3096840, 0xc000054140, 0x30b47a0, 0xc0011220c0, 0xc000e43800, 0x31, 0x2c27eda, 0x7, 0xc000658ec0, 0x32, ...)
[1]     	/home/prow/go/pkg/mod/sigs.k8s.io/cluster-api/[email protected]/framework/clusterctl/clusterctl_helpers.go:155 +0x8ab
[1]     sigs.k8s.io/cluster-api/test/e2e.ClusterctlUpgradeSpec.func2()

gets https://github.com/kubernetes-sigs/cluster-api/blob/main/test/framework/clusterctl/clusterctl_helpers.go#L153-L166
and https://github.com/kubernetes-sigs/cluster-api/blob/main/test/framework/deployment_helpers.go#L52-L73

[randomvariable]

Just ran in us-west-2, and it's working fine locally. :(

[randomvariable]

One thing to try might be to see if the e2e image upload is actually working. Will need to to look at the logs mid run, but there will be a log line along the lines of:

[1] STEP: Uploading the e2e image to s3:///capi-images.us-west-2.921092657406.oci-images/sha256:545ee55243f44f063048ea53fc99e13fc487bcd88c9b794fcf0d3e72d2b36fde
[1] STEP: Making the image public
[1] STEP: Searching for AMI: name=capa-ami-ubuntu-18.04-1.21.2*

at that point , run

aws s3 cp  s3://capi-images.us-west-2.921092657406.oci-images/sha256:545ee55243f44f063048ea53fc99e13fc487bcd88c9b794fcf0d3e72d2b36fde local.tar

Note that you'll need to change s3:/// to s3:// due to a typo in my log line.

[sedefsavas]

Thanks, will check out that.

/test pull-cluster-api-provider-aws-e2e

[randomvariable]

Yeah, the bucket is empty during the test run. I think something is deleting the S3 image. Maybe it's a lambda or some security thing deleting any object made publicly readable.

Will probably need to create an S3 bucket policy that allows members of the AWS account to read, but not set publicly readable. Might have to check if this is what is happening with k8s infra / sig testing folk.

[sedefsavas]

It is now in mid run and looks like the image is missing:
fatal error: An error occurred (404) when calling the HeadObject operation: Key "sha256:2395e936830e46b0138729b29c253f97404d382edecf4e098d268f0c7694b144" does not exist

[randomvariable]

I think I'm confident to say that the upgrades are working based on local testing, and would suggest reverting the timeout changes and the skip, and then changing Describe to PDescribe for the v1alpha3 upgrade test (rather than comment out), merge in and do a release.

Thoughts @richardcase ?. PS EKS upgrade test currently isn't possible because of #2828 dependencies.

[sedefsavas]

/test pull-cluster-api-provider-aws-e2e
Last one time before making the release.

[richardcase]

Yay @sedefsavas the e2e passed.

[sedefsavas]

/lgtm
/approve

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: sedefsavas

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 files:

• OWNERS [sedefsavas]

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

[richardcase]

I think I'm confident to say that the upgrades are working based on local testing, and would suggest reverting the timeout changes and the skip, and then changing Describe to PDescribe for the v1alpha3 upgrade test (rather than comment out), merge in and do a release.

Sounds good to me, especially as the e2e is passing. The timeout changes remain changed but all good:

/lgtm

[sedefsavas]

@richardcase These tests were already passing. The issue is with the upgrade test which I disabled in this PR.
I also confirm that it passes locally. It is not passing here because of the reason randomvariable found: the e2e image pushed to s3 is getting deleted in the prow.

[sedefsavas]

/hold cancel