Create container image repository for built images

[bowei]

We will need a gcr.io docker image repo for the built artifacts:

https://github.com/kubernetes/k8s.io/blob/master/k8s.gcr.io/README.md
https://github.com/kubernetes/test-infra/blob/master/config/jobs/image-pushing/README.md

[aojea]

/assign
xref: kubernetes/k8s.io#515

[thockin]

I am curious what artifacts you expect to build and publish?

[aojea]

@bowei only is missing the cloudbuild.yaml

https://github.com/kubernetes/test-infra/blob/master/config/jobs/image-pushing/README.md#cloudbuildyaml

and defining the artifacts #3 (comment)

[bowei]

@thockin -- conformance test suite and there are some build related images would be helpful to keep a golden image to get repeat ability.

[alexgervais]

@aojea do you know if it's possible to share a gcr.io container repo between projects? We'd like to piggyback on the work done here for the ingress-conformance images
ref. kubernetes-sigs/ingress-controller-conformance#27

[aojea]

@alexgervais do you think that is possible to use one of the kubernetes/kubernetes images used for testing?

I think that the agnhost image has the same functionality

https://github.com/kubernetes/kubernetes/tree/master/test/images/agnhost#netexec

and it's maintained by the project and published to a public registry

gcr.io/kubernetes-e2e-test-images/agnhost

[jpeach]

xref conformance test suite discussion in #20

[alexgervais]

@aojea agnhost is a very interesting project. However, it does not fit our requirements as the netexec server does not allow dumping the original request in the server's response, and injecting environment variables -- allowing us to later assert upon this data set. Also, since the behaviour of netexec is dictated by HTTP endpoints, it would not allow us to test the routing behaviour on matched and unmatched Exact or Prefix path types as all routes would be required to re-write the original request to /hostname for example.

We could of course upstream the ./tools/echoserver.go binary cmd into agnhost but I think keeping this binary and docker image closer to the ingress-conformance test suite would allow us to iterate faster.

I would also wish to publish the ingress-controller-conformance binary as a docker image for folks who would like to run the suite as part of their CI.

[aojea]

honestly don't know how this image publishing process works, we should ask sig-testing if it's possible to publish artifacts from different repos to one bucket, but doesn't seem a good solution because of the complexity of maintaining it in the long term.

Maybe you can ask for a new bucket to publish images in https://github.com/kubernetes-sigs/ingress-controller-conformance?

We will need a gcr.io docker image repo for the built artifacts:

https://github.com/kubernetes/k8s.io/blob/master/k8s.gcr.io/README.md
https://github.com/kubernetes/test-infra/blob/master/config/jobs/image-pushing/README.md

[youngnick]

So, to follow up here, there are currently two container artifacts referenced in this repo:

• k8s.gcr.io/service-apis-mkdocs:latest, this is used for all docs-related tasks and is currently built from scratch each build. Making this available globally would be useful.
• controller:latest - this is able to be created when testing the API types using the kubebuilder code. I don't think it's used anywhere yet, since the types are not functional. Obviously, it could not be made available currently because it's not namespaced.

I think the repo for the first is a guess, and is not set in stone if it needs to be pushed somewhere else. Right now, it does nothing as the result is not pushed anywhere, and there is no push functionality in docs.mk.

[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-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[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-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[fejta-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

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

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten

[youngnick]

/remove-lifecycle stale
/remove-lifecycle rotten

I think the only utility here is to make the builds/testing a bit faster by allowing reuse of the mkdocs container. Given the amount of time this has lain dormant, we can probably just close it out. Thoughts?

[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-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[hbagdi]

I think this issue now becomes important again with @cmluciano's work on validating webhook, which requires a container repository (#487).
@bowei @robscott Do you have any guidance on how to go about doing this?

[aojea]

The container repository was created, it is missing to configure the repo to automate the build
#3 (comment)

[fejta-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

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

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

[fejta-bot]

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

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

[k8s-ci-robot]

@fejta-bot: Closing this issue.

In response to this:

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

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

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.