Add tests for CAPI e2e for main branch of Kubernetes

[randomvariable]

The following e2e tests are required to provide good signal for upcoming Kubernetes releases

• Create and tear down of Kubernetes main branch
• Run conformance tests on main branch (currently done in CAPA)
• Upgrade of a cluster from (main branch - .1) to main branch.

/area testing
/priority important-critical

Related to #3564

[k8s-ci-robot]

@randomvariable: The label(s) priority/important-critical cannot be applied, because the repository doesn't have them

In response to this:

The following e2e tests are required to provide good signal for upcoming Kubernetes releases

• Create and tear down of Kubernetes main branch
• Run conformance tests on main branch (currently done in CAPA)
• Upgrade of a cluster from (main branch - .1) to main branch.

/area testing
/priority important-critical

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]

/assign

[randomvariable]

/lifecycle active

[vincepri]

/milestone v0.3.10

[vincepri]

@randomvariable The related PR has merged, should we switch the milestone, seems there is still more work in this issue

[randomvariable]

Yes, please switch. The largest chunk got merged, but there's still CAPD e2e to actually make use of the new packages.

[fabriziopandini]

/milestone v0.4.0

[randomvariable]

/lifecycle cancel
Pending #3811

[randomvariable]

/remove-lifecycle active

[fabriziopandini]

@vincepri @ncdc @naadir @jdetiber @Cecile
While #3652 provided a viable solution for all the "real" infrastructure providers, IMO downloading CI artifacts on each node, is not a viable solution for CAPD where all the machines are containers running on a single/resource-limited prow machine.

So I’m leaning towards using kind build node-image so we have a docker image with everything pre-loaded to be used for creating CAPD machines.

This step takes ~20 minutes but assuming these jobs are periodic, I don't see problems.
Opinions?

[ncdc]

👍

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-contributor-experience at kubernetes/community.
/lifecycle stale

[fabriziopandini]

/close
this is being supersede by #4043

[k8s-ci-robot]

@fabriziopandini: Closing this issue.

In response to this:

/close
this is being supersede by #4043

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.