Replatform e2e on to ginkgo v2

[randomvariable]

User Story

As a developer, I want my infrastructure cleaned up properly after an aborted test run.

Detailed Description

Looks like Ginkgo v2 will solve a lot of problems when it comes to testing on actual infrastructure. One of the main benefits is that AfterEach will be executed when tests are aborted, whereas for now only AfterSuite is. In AWS, with the current behaviour, this leaves a lot of resources behind when consuming the CAPI test framework, because we don't know the namespace the framework has created which we normally store in a node-level map for the AfterSuite to iterate through and clean up clusters.

As replatforming onto v2 (which isn't released yet) will be a breaking change, very much a v0.5.0 thing.

/milestone next

Anything else you would like to add:

[Miscellaneous information that will assist in solving the issue.]

/kind feature
/area testing
/priority important-longterm

[k8s-ci-robot]

@randomvariable: You must be a member of the kubernetes-sigs/cluster-api-maintainers GitHub team to set the milestone. If you believe you should be able to issue the /milestone command, please contact your Cluster API Maintainers and have them propose you as an additional delegate for this responsibility.

In response to this:

User Story

As a developer, I want my infrastructure cleaned up properly after an aborted test run.

Detailed Description

Looks like Ginkgo v2 will solve a lot of problems when it comes to testing on actual infrastructure. One of the main benefits is that AfterEach will be executed when tests are aborted, whereas for now only AfterSuite is. In AWS, with the current behaviour, this leaves a lot of resources behind when consuming the CAPI test framework, because we don't know the namespace the framework has created which we normally store in a node-level map for the AfterSuite to iterate through and clean up clusters.

As replatforming onto v2 (which isn't released yet) will be a breaking change, very much a v0.5.0 thing.

/milestone next

Anything else you would like to add:

[Miscellaneous information that will assist in solving the issue.]

/kind feature
/area testing
/priority important-longterm

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]

https://github.com/onsi/ginkgo/blob/v2/docs/MIGRATING_TO_V2.md if people wanna read up

[nobody4t]

This looks interesting. But from the messages of robot, some confirmation from maintainers is neccessary.
Please respond. Thanks.

[vincepri]

/milestone Next

[fabriziopandini]

We should consider also https://github.com/kubernetes-sigs/e2e-framework/

[jsturtevant]

I found the following in the e2e logs today and came here to file an issue but looks like this will cover it 👍

�[38;5;228mYou're using deprecated Ginkgo functionality:�[0m
�[38;5;228m=============================================�[0m
Ginkgo 2.0 is under active development and will introduce (a small number of) breaking changes.
To learn more, view the migration guide at �[38;5;14m�[4mhttps://github.com/onsi/ginkgo/blob/v2/docs/MIGRATING_TO_V2.md�[0m
To comment, chime in at �[38;5;14m�[4mhttps://github.com/onsi/ginkgo/issues/711�[0m

  �[38;5;11mYou are using a custom reporter.  Support for custom reporters will likely be removed in V2.  Most users were using them to generate junit or teamcity reports and this functionality will be merged into the core reporter.  In addition, Ginkgo 2.0 will support emitting a JSON-formatted report that users can then manipulate to generate custom reports.

[nobody4t]

I have spare some time to investigate the e2e-framework. Its functionality is very similar to what ginkgo/omega provides.
It is pure golang native.
But it is at very early stage. If we use this framework, we will have take care of its possible bug. It is not muture as ginkgo.
The latest ginko provides more great features which will make our test life easy.

I would like to hear more from you guys.
@fabriziopandini @vincepri @randomvariable

[fabriziopandini]

I'm +1 for investigating the e2e-framework, because AFAIK this is where Kuberenetes is going (cc @vladimirvivien), but this should be discussed in the office hours given that many providers are using the CAPI E2E test in its current form

[vincepri]

Should we separate ginkgo v2 upgrade from exploring the k8s e2e framework?

[fabriziopandini]

This is a possible way forward, but as far as I understand it won't fix #2955 (even if it will mitigate the problems that we have today for cleanup in case of errors)

[vladimirvivien]

Of course I am biased, but I think e2e-framework can help here.
I understand that currently e2e is nascent, but I believe we can get it where CAPI needs it to be.

(cc @ShwethaKumbla )

[vladimirvivien]

@vincepri and @fabriziopandini
I would love to better understand the pain points. Can you point me to somewhere in the code where you think could be done better with the new framework and I can work backward.

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle stale
• Mark this issue or PR as rotten with /lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[vladimirvivien]

/remove-lifecycle stale

Keeping this open if it's still relevant.

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle stale
• Mark this issue or PR as rotten with /lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[vladimirvivien]

/remove-lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle stale
• Mark this issue or PR as rotten with /lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[vladimirvivien]

/remove-lifecycle stale

Keeping this alive as upcoming release of e2e-framework will have features to make it easy to support testing of CRD developments including CAPI.

[fabriziopandini]

/triage accepted

As discussed in the CAPI office hours meeting, @mboersma will send an email to the list announcing the timeline for switching to Ginkgo v2.

[mboersma]

https://groups.google.com/g/kubernetes-sig-cluster-lifecycle/c/5wf5ogFUeYI