Only use the first IP address got from instance metadata. This is
because Azure CNI would setup a list of IP addresses in instance metata,
while only the first one is the Node's IP.

kubeadm: move test utilites out of /test

Remove scheduler "TestGroup" utility.

Fixing address locking in CP fake.

fix WaitForAllNodesSchedulable so it outputs logs when cluster size is small

add CNI_VERSION to build/workspace.bzl, use it to dedupe version

Plumb token and token file through rest.Config

kubeadm: remove kubeadm config print-defaults

enable secured kubelet in hack/local-up-cluster.sh

Use memory metrics from the pod cgroup for eviction ranking

Merge pull request #71597 from marc-sensenich/gh-71595/fix-cloud-controller-manager-azure-delete-nodes

Add check for cloudProvider.InstanceNotFound in Azure VMSS Instance gets

make GetStorageAccesskey func public

remove duplicate word

Signed-off-by: yuexiao-wang <wang.yuexiao@zte.com.cn>

kubeadm: properly umount dirs in /var/lib/kubelet

Update default etcd server to 3.3.10 for kubernetes 1.14

Collapse not shared NSG rules with multiple source ranges

Fix some golint errors for packages in `pkg/kubectl/cmd`

Drop the norecurse flag in  dns tests

util: Refactor Backoff to return the next step rather than sleeping

Fix typo: trus->true

Disable proxy use in http  probe

Add new reviewers for conformance

Bump the version of kubernetes-cni to 0.6.0

This util was used to fake certain aspects of apiserver behavior, such
as resource paths and JSON encoding. Our unit tests have been refactored
so they don't rely on the REST or JSON aspects of apiserver. This util
is no longer needed.

Use Node-Problem-Detector v0.6.0 on GCI

dnsmasq 2.79 introduced a change to respond to all norecurse queries with ServFail.
This is to prevent cache snooping where an adversary can figure out if a particular hostname has been looked up or not.
These tests do not need the norecurse flag, hence removing it.

Issue 71614: Protect log message maps

Allows consumers to use Backoff as a generator rather than have to
call ExponentialBackoff

Update changelogs for CVE-2018-1002105

'kubeadm reset' uses incorrect way of unmounting /var/lib/kubelet
directories. It queries /proc/mounts for /var/lib/kubelet mount point.
If /var/lib/kubelet directory is also mounted it makes 'kubelet reset'
to unmount it too, which is incorrect. It also makes it fail as it
can't unmount /var/lib/kubelet before unmounting mounts inside it.

Fixed by querying /var/lib/kubelet/ instead of /var/lib/kubelet.
This should exclude /var/lib/kubelet from the query results even if
it's mounted.

Fixes: kubernetes/kubeadm#1294

Signed-off-by: Rostislav M. Georgiev <rostislavg@vmware.com>