Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add E2E tests #233

Closed
terrytangyuan opened this issue Apr 20, 2020 · 16 comments
Closed

Add E2E tests #233

terrytangyuan opened this issue Apr 20, 2020 · 16 comments
Assignees

Comments

@terrytangyuan
Copy link
Member

Part of #217.

  • E2E tests should be run on presubmit, postsubmit, and periodically
  • Tests should cover deployment (kustomize packages) as well as the application functionality
@NikeNano
Copy link
Member

NikeNano commented Apr 21, 2020

I am looking into this, but may not be super fast. Need to catch up on prow and how the test infrastructure for Kubeflow works.

Do I understand it correct that prow is not used atm, since the prow config is empty @terrytangyuan ?

@NikeNano
Copy link
Member

/assign

@terrytangyuan
Copy link
Member Author

Thanks! Yes that’s correct.

@NikeNano
Copy link
Member

I have played around with the test infra a bite. Do we have any thought about how we like the e2e test to look like? Should we build it based upon one of the examples? If so I suggest that we use the v1alpha2/tensorflow-benchmark.

@terrytangyuan
Copy link
Member Author

Yes that would be a good example to use for end2end test.

@terrytangyuan
Copy link
Member Author

Actually v1 controller is the one we want to graduate so use this example instead: https://github.com/kubeflow/mpi-operator/blob/master/examples/v1/tensorflow-benchmarks.yaml

@naveensrinivasan
Copy link
Member

@terrytangyuan How about doing e2e using something like k3d or kind? This will give confidence for someone to run locally. Similar to this kubeflow/kubeflow#5400

@terrytangyuan
Copy link
Member Author

See #296 for an alternative approach to leverage AWS's test infra.

@terrytangyuan
Copy link
Member Author

cc @kubeflow/wg-training-leads @PatrickXYS

@PatrickXYS
Copy link
Member

FYI, you can check migration doc here https://github.com/kubeflow/testing/blob/master/aws/Doc/onboarding.md

@naveensrinivasan
Copy link
Member

I agree that the end goal is to use the real infrastructure.

As a contributor I would like to run functional test prior submitting the PR. Also I would like to run partial set of tests based on the changes.

Can you explain what is the disadvantages of using something in memory and also have an option to use the actual infrastructure based on a flag?

This gives significant advantages of having a short dev/test cycle.

@naveensrinivasan
Copy link
Member

naveensrinivasan commented Dec 25, 2020

FYI, you can check migration doc here https://github.com/kubeflow/testing/blob/master/aws/Doc/onboarding.md

@PatrickXYS How do I get access to the repo aws-kf-ci-bot?

I would like to start working on the e2e tests.

@PatrickXYS
Copy link
Member

@PatrickXYS How do I get access to the repo aws-kf-ci-bot?

You won't, you should follow the doc to proceed

@alculquicondor
Copy link
Collaborator

I started to look into this (along with integration tests). Did anyone make any progress?
I'm thinking of using this package https://pkg.go.dev/sigs.k8s.io/controller-runtime/pkg/envtest

@alculquicondor
Copy link
Collaborator

alculquicondor commented Aug 16, 2021

With #399 and #403 merged, I think we should be good to close this.

@terrytangyuan wdyt?

@terrytangyuan
Copy link
Member Author

terrytangyuan commented Aug 16, 2021

Sure. Thank you!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

5 participants