cloudLabels for API ELB

[hridyeshpant]

It seems cloudLabels is not putting TAG on API ELB's.
kops create cluster --v=3 --state=s3://ewe-kubernetes-dev --cloud=aws
--name=kubetnets.us-west-2.dev.expedia.com
--ssh-public-key=~/.ssh/kube_aws_rsa.pub
--master-size=t2.medium
--cloud-labels="Team=CAT"
--master-zones=us-west-2c,us-west-2a,us-west-2b
--network-cidr=10.4.0.0/16
--vpc=vpc-030e8f64
--node-count=3
--node-size=t2.micro
--topology private
--networking weave
--zones=us-west-2c,us-west-2a,us-west-2b
--channel=alpha

Result : k8 API ELB's dont have AWS TAG mentioned in cloud-labels.

[richarvey]

I can confirm this is the case, labels are not applied to the ELB, so if you have a AWS janitor to clear up untagged resources this is an issue.

[estebancrw]

I'm also having this issue. Any way to work around it at the moment or any plans into adding this use case in the future?
This can be done in k8s since v1.7.0 with annotations service.beta.kubernetes.io/aws-load-balancer-additional-resource-tags, but I'm not sure how to do this with the ELB from kops

[geojaz]

My understanding is that cloud labels are associated with InstanceGroups, not all AWS resources. In this case, I think kops is acting as intending.

It looks like you'd like to be able to assign a "global" tag to all resources in your cluster? This sounds like a reasonable thing to do and I would be open to thinking about implementation.

[chrislovecnm]

So we already have global tags, KubernetesCluster for instance. But not sure if we should use it in the elb, as I am not certain how the cloud provider will react ;)

Let me know Eric

[estebancrw]

@geojaz That is exactly my use-case. Some organizations have created "janitors" that delete AWS resources not tagged as they specify. This "janitors" tend to be for enforcing their tagging policies so they can bill correctly their own business units and budget accordingly.

@chrislovecnm would be an awesome feature to be able to extend the CloudLabels into global tags.

[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

[brantburnett]

/remove-lifecycle stale

This is also an issue for my company, for the same reasons as mentioned above surrounding billing

[jcardoso-bv]

Also an issue for my company (billing).

[tonidas]

That's really an issue!

[flamingspaz]

Has anyone found a workaround for this, other than manually applying the tags?

[lvicentesanchez]

I'm also suffering from this problem

[wingyplus]

I just sent a patch to fix this problem. Needs someone to help me test this patch. Suggestions are welcome. :)

[chrisz100]

/reopen

Fix seems to have gotten reverted

[k8s-ci-robot]

@chrisz100: Reopened this issue.

In response to this:

/reopen

Fix seems to have gotten reverted

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.