Sets min kube-dns replica number to 2

[MrHohn]

From #40063.

Setting the min value to 2 in dns-horizontal-autoscaler to mitigate kube-dns single point failure. The downside is we will have 2 replicas even if there is only 1 node, though this is still a better setup.

@thockin @bowei

[k8s-reviewable]

This change is 

[k8s-ci-robot]

Jenkins GCI GKE smoke e2e failed for commit 03efda4. Full PR test history.

cc @MrHohn, your PR dashboard

The magic incantation to run this job again is @k8s-bot gci gke e2e test this. Please help us cut down flakes by linking to an open flake issue when you hit one in your PR.

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. I understand the commands that are listed here.

[k8s-ci-robot]

Jenkins GKE smoke e2e failed for commit 03efda4. Full PR test history.

cc @MrHohn, your PR dashboard

The magic incantation to run this job again is @k8s-bot cvm gke e2e test this. Please help us cut down flakes by linking to an open flake issue when you hit one in your PR.

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. I understand the commands that are listed here.

[k8s-ci-robot]

Jenkins Kubemark GCE e2e failed for commit 03efda4. Full PR test history.

cc @MrHohn, your PR dashboard

The magic incantation to run this job again is @k8s-bot kubemark e2e test this. Please help us cut down flakes by linking to an open flake issue when you hit one in your PR.

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. I understand the commands that are listed here.

[thockin]

We don't really want a min of 2 - on a 1 node cluster that will run DNS twice. We used to have different configs for node count and CPU count - did we eliminate that? That was what let us have a bi-modal config - a ladder for nodes and a scaler for cores.

[MrHohn]

I see your point now. But sadly current autoscaler is unable to support that --- it either use linear mode for both nodes and cores or use ladder mode for both.

[thockin]

Open to suggestions to get the right semantics... I feel like we had this correct before.

…

On Sun, Jan 22, 2017 at 6:47 PM, Zihong Zheng ***@***.***> wrote: I see your point now. But sadly current autoscaler is unable to support that --- it either use linear mode for both nodes and cores or use ladder mode for both. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#40281 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFVgVME2qqnIG3bp0eGHp9J5hEZnoiOkks5rVBS0gaJpZM4LqiOs> .

[k8s-github-robot]

[APPROVALNOTIFIER] Needs approval from an approver in each of these OWNERS Files:

• cluster/addons/dns-horizontal-autoscaler/OWNERS

We suggest the following people:
cc @bowei
You can indicate your approval by writing /approve in a comment
You can cancel your approval by writing /approve cancel in a comment

[MrHohn]

Close it for now.