Boskos should have an automated push/bump/deploy like Prow

[Katharine]

Doing things manually invites mistakes!

/assign @fejta

[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

[Katharine]

/remove-lifecycle stale

it automatically pushes and bumps, but still does not automatically deploy.

[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

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

[stevekuznetsov]

/remove-lifecycle rotten
/reopen

We're deploying it automatically on the OpenShift side, so this is definitely possible.

[k8s-ci-robot]

@stevekuznetsov: Reopened this issue.

In response to this:

/remove-lifecycle rotten
/reopen

We're deploying it automatically on the OpenShift side, so this is definitely possible.

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.

[cjwagner]

/kind oncall-hotlist

The boskos images are currently running at version v20190621-ff01381 even though there have been many changes since then: https://github.com/kubernetes/test-infra/commits/master/boskos

@michelle192837 (current oncall) Could you please publish and deploy the latest boskos images when you get the chance (or make it automatic)?

[fejta]

This is the latest image:

$ gcloud container images list-tags gcr.io/k8s-testimages/boskos
9abcb3ce0bea  v20190621-ff01381                                         2019-06-21T09:04:24

The prow bump already autobumps this to the latest image whenever it changes.

[fejta]

We also have a job to push images whenever they change:

test-infra/config/jobs/kubernetes/test-infra/test-infra-trusted.yaml

Lines 178 to 204 in 8d07c82

	- name: post-test-infra-push-boskos
	cluster: test-infra-trusted
	run_if_changed: '^(boskos\/|maintenance\/aws-janitor)'
	annotations:
	testgrid-dashboards: "sig-testing-images"
	testgrid-tab-name: "boskos"
	testgrid-alert-email: kubernetes-sig-testing-alerts@googlegroups.com, k8s-infra-oncall@google.com
	testgrid-num-failures-to-alert: '1'
	description: builds and pushes boskos images
	decorate: true
	branches:
	- master
	spec:
	containers:
	- image: gcr.io/k8s-testimages/bazelbuild:v20190916-ec59af8-0.29.1 # whatever image you use here must have bash 4.4+
	command:
	- boskos/push.sh
	env:
	- name: GOOGLE_APPLICATION_CREDENTIALS
	value: /creds/service-account.json
	volumeMounts:
	- name: creds
	mountPath: /creds
	volumes:
	- name: creds
	secret:
	secretName: pusher-service-account

[cjwagner]

We also have a job to push images whenever they change

I don't think this part is working right. There have been many changes to boskos since June 21, but that is the most recent image we've published: https://github.com/kubernetes/test-infra/commits/master/boskos

If you look at the post submit job statuses on those commits it does look like the post-test-infra-push-boskos job is running, but the images it says it is publishing are not actually being published. e.g. https://prow.k8s.io/view/gcs/kubernetes-jenkins/logs/post-test-infra-push-boskos/1173991256389324801 (should have published tag v20190917-d326d2a31.

[cjwagner]

Ooooh, looks like we have published boskos images to both gcr.io/k8s-testimages and gcr.io/k8s-prow. Our deployments are all referencing gcr.io/k8s-testimages still, but we only push to gcr.io/k8s-prow now so autobump isn't finding the newer images.

[michelle192837]

I believe this is fixed now, and automatically bumps when we do the Prow bump. (Reopen if I'm wrong :P )