WIP update go mod for test

[zw0610]

This is to test kubeflow/common#173.. Please don't review/approve it.

/hold

[google-oss-prow]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: zw0610
To complete the pull request process, please assign jeffwan after the PR has been reviewed.
You can assign the PR to them by writing /assign @jeffwan in a comment when ready.

The full list of commands accepted by this bot can be found here.

Needs approval from an approver in each of these files:

• OWNERS

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

[coveralls]

Pull Request Test Coverage Report for Build 1486667268

• 0 of 2 (0.0%) changed or added relevant lines in 1 file are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.03%) to 24.099%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
pkg/controller.v1/tensorflow/tfjob_controller.go	0	2	0.0%

Totals
Change from base Build 1483079971:	0.03%
Covered Lines:	1197
Relevant Lines:	4967

---

💛 - Coveralls

[zw0610]

/retest

[Jeffwan]

Logs are no longer available there. I can help rerun the tests and help check the issue. Let me also write a debug doc to show how to kick off those e2e tests in local envs

[zw0610]

Logs are no longer available there. I can help rerun the tests and help check the issue. Let me also write a debug doc to show how to kick off those e2e tests in local envs

All I can see is timeout error for these 3 failed test. Please retest it and help check the issue. Thanks!

[Jeffwan]

em.. Seems I need to change another dev machine to do the test. git connection is really unstable in CN.

ks-13 env add test-env-2f78 --api-spec=version:v1.13.0 --namespace=kubeflow
INFO Using context "docker-desktop" from kubeconfig file "/Users/jiaxin/.kube/config"
INFO Creating environment "test-env-2f78" with namespace "kubeflow", pointing to "version:v1.13.0" cluster at address "https://kubernetes.docker.internal:6443"
ERROR Get https://raw.githubusercontent.com/kubernetes/kubernetes/v1.13.0/api/openapi-spec/swagger.json: dial tcp [::]:443: connect: connection refused

[Jeffwan]

/test kubeflow-training-operator-presubmit

[Jeffwan]

I manually run CleanPodPolicyTests and PodNamesValidationTest in my environment and it succeed. Logs and logics look good to me. I rerun the CI jobs to check if there's other problem (I have not looked at original failure messages)

<testsuite failures="0" tests="1" time="131.18602418899536"><testcase classname="CleanPodPolicyTests" name="cleanpod-policy-tests-v1" time="131.18602418899536" /></testsuite>
<testsuite failures="0" tests="1" time="40.802759408950806"><testcase classname="PodNamesValidationTest" name="pod-names-validation-tests-v1" time="40.802759408950806" /></testsuite>

[Jeffwan]

service:

pod:

I think I found the problem. the service label and selector are still using upper case so that's the reason it can not find the underneath pods which makes real testing jobs failed.

The reason that CleanPodPolicy works is because it only test policy instead of real workloads, they are using a fake image ubuntu.

training-operator/test/workflows/components/clean_pod_running_v1.jsonnet

Line 33 in 654540b

image: "ubuntu",

@zw0610 please help update kubeflow/common and let's give another try

[zw0610]

/close as kubeflow/common#173 is merged.