f9a45e6983
Automatic merge from submit-queue add default flannel configuration for cni I added a default flannel configuration for cni. So hyperkube can be started with: ``` --network-plugin=kubenet ``` or with flannel: ``` --network-plugin=cni --network-plugin-dir=/etc/cni/net.d ``` Requires update of cni binaries to 0.3.0 or later to work. See also https://github.com/kubernetes/kubernetes/issues/27603 This PR is intended to support multi-node Hyperkube https://github.com/kubernetes/kube-deploy/pull/115 CC @zreigz []()
hyperkube
hyperkube is an all-in-one binary for the Kubernetes server components
Also, it's very easy to run this hyperkube setup dockerized.
See http://kubernetes.io/docs/getting-started-guides/docker/ for up-to-date commands.
hyperkube is built for multiple architectures and pushed automatically on every release.
How to release by hand
# First, build the binaries
$ build/run.sh make cross
# Build for linux/amd64 (default)
$ make push VERSION={target_version} ARCH=amd64
# ---> gcr.io/google_containers/hyperkube-amd64:VERSION
# ---> gcr.io/google_containers/hyperkube:VERSION (image with backwards-compatible naming)
$ make push VERSION={target_version} ARCH=arm
# ---> gcr.io/google_containers/hyperkube-arm:VERSION
$ make push VERSION={target_version} ARCH=arm64
# ---> gcr.io/google_containers/hyperkube-arm64:VERSION
$ make push VERSION={target_version} ARCH=ppc64le
# ---> gcr.io/google_containers/hyperkube-ppc64le:VERSION
If you don't want to push the images, run make or make build instead
