Change constants CoreDNSImageName

[thdrnsdk]

What happened:

When installing using kubeadm, only the image of coredns has a different depth.
This is very inconvenient when installing using a private registry.

What you expected to happen:

$ kubeadm config images list
k8s.gcr.io/coredns/coredns:v1.8.0

change to

k8s.gcr.io/coredns:v1.8.0

How to reproduce it (as minimally and precisely as possible):

replace constants in

kubernetes/cmd/kubeadm/app/constants/constants.go

Line 331 in 6a7572e

CoreDNSImageName = "coredns/coredns"

Anything else we need to know?:

need to push again to the gcr repository .

[k8s-ci-robot]

@thdrnsdk: This issue is currently awaiting triage.

If a SIG or subproject determines this is a relevant issue, they will accept it by applying the triage/accepted label and provide further guidance.

The triage/accepted label can be added by org members by writing /triage accepted in a comment.

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.

[thdrnsdk]

/sig cluster-lifecycle

[pacoxu]

#99321 (comment)

From v1.7.1, the image promotion is no longer done in the legacy way

v1.8.0 should be available under k8s.gcr.io/coredns/coredns:v1.8.0

[pacoxu]

/area dns

[thdrnsdk]

The image depth only changes when using coredns images.
In case of building and using private Image Registry, it is strange to have subpath repository level for only coredns.

k8s.gcr.io/kube-apiserver:v1.21.0
k8s.gcr.io/kube-controller-manager:v1.21.0
k8s.gcr.io/kube-scheduler:v1.21.0
k8s.gcr.io/kube-proxy:v1.21.0
k8s.gcr.io/pause:3.4.1
k8s.gcr.io/etcd:3.4.13-0
k8s.gcr.io/coredns/coredns:v1.8.0

[pacoxu]

/cc @rajansandeep

[wangyysde]

@thdrnsdk
Yes, I am agree with you. It is very strange that only the path of coredns is /coredns/xxxx ,the others is /xxx. @fabriziopandini @SataQiu Can we change the path of coredns to /xxx too?

[fabriziopandini]

This is not really up to Kubeadm.
kubeadm simply adapts to change that happens somewhere else, so changing the constant in kubeadm won't fix the image name; instead, it will make v1.21 clusters to fail.

Please refert to kubernetes-sigs/cluster-api#4463 (comment) for a detailed analysis of available images for CoreDNS; Also, please note that changes to the image name have side effects in other systems as well, so, if possible, I suggest to avoid changing this again...

[dims]

+1 to what @fabriziopandini said.

In all honesty, more things will go to the format that coredns is following over time. So we better get used to it.

[neolit123]

the depth cannot be changed and you must adapt your local registry.

cc @chrisohaver @rajansandeep
since the change of path for the image and introduction of v created similar complains from users.

/close

[k8s-ci-robot]

@neolit123: Closing this issue.

In response to this:

the depth cannot be changed and you must adapt your local registry.

cc @chrisohaver @rajansandeep
since the change of path for the image and introduction of v created similar complains from users.

/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.