Got 403 when trying to get a release for release-1.4 about release

[david-mcmahon]

Can you attach a full log?

[pwittrock]

Yes I will do that. I will need to scp them from my machine this afternoon.

[pwittrock]

anago.log

[pwittrock]

Looks like the issue is pushing to gcr.io. If I run the script as the production user then it tells me to switch to my personal gcloud user, but then it fails to push to gcr. Is this because I am running in mock mode?

/usr/local/google-cloud-sdk/bin/gcloud docker push gcr.io/kubernetes-release-test/kube-proxy:v1.4.0-beta.1
The push refers to a repository [gcr.io/kubernetes-release-test/kube-proxy](len: 1)
91dc026257fb: Preparing
Post https://gcr.io/v2/kubernetes-release-test/kube-proxy/blobs/uploads/: token auth attempt for registry: https://gcr.io/v2/token?account=oauth2accesstoken&scope=repository%3Akubernetes-release-test%2Fkube-proxy%3Apush%2Cpull&service=gcr.io request failed with status: 403 Forbidden

[pwittrock]

It looks like the production user only gets activated when running in nomock, and I do not have permissions to write to the mock registry

[[ "$registry" =~ gcr.io/ ]] && docker_push_cmd=("$GCLOUD" "docker")

  # Activate credentials for the [email protected]
  [[ "$registry" == "gcr.io/google_containers" ]] \
   && logrun $GCLOUD config set account [email protected]

  logecho
  logecho "Send docker containers to $registry..."

Log Line

anago::release::docker::release(): Send docker containers to gcr.io/kubernetes-release-test...

So it is just trying to push as me. I am going to try to run in nomock mode to see if solves the problem.

[pwittrock]

I tried updating the script to always change the user and pushing the images still fails...

[david-mcmahon]

Looks like you need permission to gcr.io/kubernetes-release-test.
There's also a bug that this isn't checked in the prereqs.

[pwittrock]

How do I get that :P

[fejta-bot]

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

Prevent issues from auto-closing with an /lifecycle frozen comment.

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

[david-mcmahon]

Auth issue, worked out at the time.
Since then a check has been added to validate access to kubernetes-release-test as well.