v0.3.0

• [enhancement] Update containerd to 1.4.1 and CNI plugins 0.8.7 #28
• [enhancement] Support running on aarch64 arch #26
• [enhancement] Support kubeadm init options #25
• [enhancement] Add 'cluster config-template' command to generate the default cluster configuration #22
• [enhancement] Remove admin node support #21
• [enhancement] Add image command to show supported rootfs/kernel images #20
• [enhancement] Add ubuntu:20.04 LTS image #19
• [enhancement] Move pre-builtin rootfs and kernel OCI images to github container registry #18
• [enhancement] Support declarative configuration to deploy cluster #9
• [enhancement] Support updating installed prerequisites #8
• [enhancement] Support customized kubeadm control plane configurations #7

Notes

Because of upgrading containerd to 1.4.1 and CNI plugins to 0.8.7, please remember to update your installed prerequisites via the below command.

kubefire install

v0.2.1

• [bug] Can't deploy 1.19.1 via kubeadm #17

Notes

Please update to this patch version to fix #17 as below.

➜  kubectl get nodes                        
NAME               STATUS   ROLES    AGE     VERSION
8cbcdfcf912d5e2c   Ready    master   8m58s   v1.19.1

➜  kubefire cluster get demo-kubeadm -o yaml
name: demo-kubeadm
spec:
  name: demo-kubeadm
  bootstrapper: kubeadm
  pubkey: /home/davidko/.kubefire/clusters/demo-kubeadm/key.pub
  prikey: /home/davidko/.kubefire/clusters/demo-kubeadm/key
  version: v1.19.1
  image: innobead/kubefire-opensuse-leap:15.2
  kernel_image: innobead/kubefire-kernel-4.19.125-amd64:latest
  kernel_args: console=ttyS0 reboot=k panic=1 pci=off ip=dhcp security=apparmor apparmor=1
  admin:
    count: 0
    memory: 512MB
    cpus: 1
    disk_size: 2GB
  master:
    count: 1
    memory: 2GB
    cpus: 2
    disk_size: 10GB
  worker:
    count: 0
    memory: 2GB
    cpus: 2
    disk_size: 10GB
  extra_options: ""
  deployed: true
nodes:
- name: demo-kubeadm-master-1
  spec:
    count: 0
    memory: 2.0 GB
    cpus: 2
    disk_size: 10.0 GB
  status:
    running: true
    ipaddresses: 10.62.0.105
    image: docker.io/innobead/kubefire-opensuse-leap@sha256:a25452ac19cd5f621eabc58bf68402cf09f1f76b50116412e646bfe3dc5d6653
    kernel: docker.io/innobead/kubefire-kernel-4.19.125-amd64@sha256:5ff327bb76d96c07493fdf0540f27d3aacf0ed6ed19cbd372282ed5af094d0d8

➜  kubefire version
Version: v0.2.1
Build: 9eef28e-20200909182753

➜  kubefire info -b
KUBEADM                 	K3S                      	SKUBA         
v1.19.1,v1.18.8,v1.17.11	v1.16.15,v1.17.11,v1.18.8	v1.4.1,v1.3.5	

v0.2.0 - updated

• [enhancement] Add complete node/cluster management commands #10
• [enhancement] Support selective Kubernetes version to create different versions of clusters #5
• [enhancement] Have cluster running after the nodes restart #3
• [enhancement] Update containerd to 1.4.0
• [enhancement] Introduce kubefire kubeconfig get/download commands
• [enhancement] Introduce kubefire info -b to show supported K8s/K3s versions
• [bug] Fix failed single-node deployment by skuba
• [bug] Invalid version ctr tool used when deploying 1.19.0 with kubeadm #14

Notes

• Because of upgrading containerd to 1.4.0, please remember to update your installed prerequisites via the below command.

kubefire install

• 1.19.0 (bootstrapper=kubeadm) introduced

➜  kubefire cluster create demo
....
➜  kubectl get nodes
NAME               STATUS   ROLES    AGE     VERSION
0ddc5695aab517ca   Ready    master   2m22s   v1.19.0

Deprecations

• kubefire cluster download was deprecated and replaced with kubefire kubeconfig get/download to support printing out to the stand output or downloading the file as before.

v0.1.0

This the first release of Kubefire 🔥

Functions

• Kubefire prerequisites management
  • Kubernetes/K3s v1.18.8 supported
  • install/uninstall prerequisites
• Cluster deployment by kubeadm, k3s, skuba
• Cluster management (local host)
  • create/delete/get/list cluster
  • start/stop cluster with the same allocated IPs (supported by host-local-rev CNI plugin)
  • download cluster kubeconfig admin.conf
  • print out KUBECONFIG environment variable
• Node access via SSH

Quickstart

Running below commands is to quickly have a cluster deployed by kubeadm running in minutes.

kubefire install
kubefire cluster create demo