文章目录
1. 简介
K3s 是一个轻量级的 Kubernetes 发行版,在 2020 年统计的 K3s 下载量中,K3s 的全球下载量已经超过 100 万次,每周平均被安装超过 2 万次,其中 30%的下载量来自于中国。在国内已经有许多用户将 K3s 应用到了各种边缘计算和物联网设备中,同时也被广泛应用于智能工厂部署的生产线机器人和一些世界上最大型的风力发电厂当中。
针对生产环境下的 K3s,一个不可逾越的问题就是离线安装。在你的离线环境需要准备以下 3 个组件:
- K3s 的安装脚本
- K3s 的二进制文件
- K3s 依赖的镜像
通过K3s Release页面(https://github.com/k3s-io/k3s/releases )下载二进制文件与镜像,如果在国内使用,推荐从 http://mirror.cnrancher.com 获得这些组件。
离线安装的重点在于K3s 依赖的镜像部分,因为 K3s 的"安装脚本"和"二进制文件"只需要下载到对应目录,然后赋予相应的权限即可,非常简单。但K3s 依赖的镜像的安装方式取决于你使用的是手动部署镜像还是私有镜像仓库,也取决于容器运行时使用的是
containerd
还是
docker
。
针对不同的组合形式,可以分为以下几种形式来实现离线安装:
Containerd
+ 手动部署镜像方式Docker
+ 手动部署镜像方式Containerd
+ 私有镜像仓库方式Docker
+ 私有镜像仓库方式
2. Docker + 手动部署镜像方式
假设你已经将同一版本的 K3s 的安装脚本(
k3s-install.sh
)、K3s 的二进制文件(
k3s
)、K3s 依赖的镜像(
k3s-airgap-images-amd64.tar
)下载到了/root目录下。
与 containerd 不同,使用 docker 作为容器运行时,启动 K3s 不会导入
/var/lib/rancher/k3s/agent/images/
目录下的镜像。所以在启动 K3s 之前我们需要将 K3s 依赖的镜像手动导入到 docker 镜像列表中。
$ ls
k3s k3s-airgap-images-amd64.tar k3s-install.sh
$ cp k3s /usr/local/bin/
2.1 安装docker
- docker 安装
2.2 导入镜像
$ docker load -i k3s-airgap-images-amd64.tar
4fc242d58285: Loading layer [==================================================>]5.855 MB/5.855 MB
43ca363a6184: Loading layer [==================================================>]17.5 MB/17.5 MB
ac85d27cf099: Loading layer [==================================================>]90.36 MB/90.36 MB
f95c9a53e804: Loading layer [==================================================>]125.6 MB/125.6 MB
Loaded image: rancher/klipper-helm:v0.7.3-build20220613
eb4bde6b29a6: Loading layer [==================================================>]5.876 MB/5.876 MB
b73f2911df5c: Loading layer [==================================================>]2.625 MB/2.625 MB
681d12cb8ee1: Loading layer [==================================================>]3.584 kB/3.584 kB
Loaded image: rancher/klipper-lb:v0.3.5
8d3ac3489996: Loading layer [==================================================>]5.866 MB/5.866 MB
2d6fddee6202: Loading layer [==================================================>]29.39 MB/29.39 MB
Loaded image: rancher/local-path-provisioner:v0.0.21
256bc5c338a6: Loading layer [==================================================>]336.4 kB/336.4 kB
5cc7f3ffa7e3: Loading layer [==================================================>]49.33 MB/49.33 MB
Loaded image: rancher/mirrored-coredns-coredns:1.9.1
0b16ab2571f4: Loading layer [==================================================>]1.459 MB/1.459 MB
Loaded image: rancher/mirrored-library-busybox:1.34.1
34d5ebaa5410: Loading layer [==================================================>]5.866 MB/5.866 MB
694ee27c6a0b: Loading layer [==================================================>]2.852 MB/2.852 MB
e32614d8591c: Loading layer [==================================================>]99.74 MB/99.74 MB
44aad00a5195: Loading layer [==================================================>]2.048 kB/2.048 kB
Loaded image: rancher/mirrored-library-traefik:2.9.1
5b1fa8e3e100: Loading layer [==================================================>]3.697 MB/3.697 MB
3dc34f14eb83: Loading layer [==================================================>]66.43 MB/66.43 MB
Loaded image: rancher/mirrored-metrics-server:v0.6.1
1021ef88c797: Loading layer [==================================================>]684.5 kB/684.5 kB
Loaded image: rancher/mirrored-pause:3.6
$ docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
rancher/mirrored-library-traefik 2.9.1 e6de8578b238 4 weeks ago 107 MB
rancher/mirrored-library-busybox 1.34.1 ff4a8eb070e1 4 weeks ago 1.24 MB
rancher/klipper-helm v0.7.3-build20220613 38b3b9ad736a 4 months ago 239 MB
rancher/mirrored-coredns-coredns 1.9.1 99376d8f35e0 7 months ago 49.5 MB
rancher/mirrored-metrics-server v0.6.1 e57a417f15d3 8 months ago 68.8 MB
rancher/klipper-lb v0.3.5 dbd43b6716a0 9 months ago 8.09 MB
rancher/local-path-provisioner v0.0.21 fb9b574e03c3 10 months ago 35 MB
rancher/mirrored-pause 3.6 6270bb605e12 14 months ago 683 kB
2.3 安装 k3s
$ INSTALL_K3S_SKIP_DOWNLOAD=true INSTALL_K3S_EXEC='--docker' /root/k3s-install.sh
[INFO] Skipping k3s download and verify
[INFO] Skipping installation of SELinux RPM
[INFO] Skipping /usr/local/bin/kubectl symlink to k3s, command exists inPATH at /usr/bin/kubectl
[INFO] Skipping /usr/local/bin/crictl symlink to k3s, command exists inPATH at /usr/bin/crictl
[INFO] Skipping /usr/local/bin/ctr symlink to k3s, command exists inPATH at /usr/bin/ctr
[INFO] Creating killall script /usr/local/bin/k3s-killall.sh
[INFO] Creating uninstall script /usr/local/bin/k3s-uninstall.sh
[INFO] env: Creating environment file /etc/systemd/system/k3s.service.env
[INFO] systemd: Creating servicefile /etc/systemd/system/k3s.service
[INFO] systemd: Enabling k3s unit
Created symlink from /etc/systemd/system/multi-user.target.wants/k3s.service to /etc/systemd/system/k3s.service.
[INFO] systemd: Starting k3s
2.4 查看
k3s kubectl get pods -A
NAMESPACE NAME READY STATUS RESTARTS AGE
kube-system local-path-provisioner-5b5579c644-fmq74 1/1 Running 0 59s
kube-system coredns-75fc8f8fff-nhn4s 1/1 Running 0 59s
kube-system helm-install-traefik-crd-pksg7 0/1 Completed 0 60s
kube-system metrics-server-5c8978b444-7tmd9 1/1 Running 0 59s
kube-system svclb-traefik-db846a66-mq5wk 2/2 Running 0 32s
kube-system traefik-9c6dc6686-fpj6s 1/1 Running 0 32s
kube-system helm-install-traefik-66zjk 0/1 Completed 2 60s
把 master 节点机器上的
/etc/rancher/k3s/k3s.yaml
文件内容写入到
~/.kube/config
文件,不要忘记修改 server 地址改为 master 节点地址:
$ kubectl get node
Unable to connect to the server: x509: certificate signed by unknown authority
$ cp /root/.kube/config /tmp/kubeconfig
$ cp /etc/rancher/k3s/k3s.yaml /root/.kube/config
cp: overwrite ‘/root/.kube/config’? y
$ vim /root/.kube/config
apiVersion: v1
clusters:
- cluster:
certificate-authority-data: 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
server: https://k3s_master:6443 #修改127.0.0.1为master hostname(k3s_master)
name: default
contexts:
......
$ kubectl get ns
NAME STATUS AGE
default Active 4m45s
kube-system Active 4m45s
kube-public Active 4m45s
kube-node-lease Active 4m45s
$ kubectl get node
NAME STATUS ROLES AGE VERSION
k3smaster Ready control-plane,master 4m50s v1.25.3+k3s1
3. Containerd + 手动部署镜像方式
假设你已经将同一版本的 K3s 的安装脚本(k3s-install.sh)、K3s 的二进制文件(k3s)、K3s 依赖的镜像(
k3s-airgap-images-amd64.tar
)下载到了/root目录下。
如果你使用的容器运行时为
containerd
,在启动 K3s 时,它会检查
/var/lib/rancher/k3s/agent/images/
是否存在可用的镜像压缩包,如果存在,就将该镜像导入到containerd 镜像列表中。所以我们只需要下载 K3s 依赖的镜像到
/var/lib/rancher/k3s/agent/images/
目录,然后启动 K3s 即可。
3.1 导入镜像到 containerd 镜像列表
sudomkdir -p /var/lib/rancher/k3s/agent/images/
sudocp /root/k3s-airgap-images-amd64.tar /var/lib/rancher/k3s/agent/images/
3.2 授予可执行权限
sudochmod a+x /root/k3s /root/k3s-install.sh
sudocp /root/k3s /usr/local/bin/
3.3 安装 K3s
INSTALL_K3S_SKIP_DOWNLOAD=true /root/k3s-install.sh
稍等片刻,即可查看到 K3s 已经成功启动:
$ crictl images
IMAGE TAG IMAGE ID SIZE
docker.io/rancher/coredns-coredns 1.8.0 296a6d5035e2d 42.6MB
docker.io/rancher/klipper-helm v0.3.2 4be09ab862d40 146MB
docker.io/rancher/klipper-lb v0.1.2 897ce3c5fc8ff 6.46MB
docker.io/rancher/library-busybox 1.31.1 1c35c44120825 1.44MB
docker.io/rancher/library-traefik 1.7.19 aa764f7db3051 86.6MB
docker.io/rancher/local-path-provisioner v0.0.14 e422121c9c5f9 42MB
docker.io/rancher/metrics-server v0.3.6 9dd718864ce61 41.2MB
docker.io/rancher/pause 3.1 da86e6ba6ca19 746kB
$ kubectl get pods -A
NAMESPACE NAME READY STATUS RESTARTS AGE
kube-system local-path-provisioner-7c458769fb-zdg9z 1/1 Running 0 38s
kube-system coredns-854c77959c-696gk 1/1 Running 0 38s
kube-system metrics-server-86cbb8457f-hs6vw 1/1 Running 0 38s
kube-system helm-install-traefik-4pgcr 0/1 Completed 0 38s
kube-system svclb-traefik-bq7wl 2/2 Running 0 17s
kube-system traefik-6f9cbd9bd4-jccd7 1/1 Running 0 17s
4. Containerd + 私有镜像仓库方式
假设你已经将同一版本的 K3s 的安装脚本(k3s-install.sh)、K3s 的二进制文件(k3s)下载到了/root目录下。并且 K3s 所需要的镜像已经上传到了镜像仓库(本例的镜像仓库地址为:http://192.168.64.44:5000)。K3s 所需的镜像列表可以从 K3s Release页面的
k3s-images.txt
获得。
4.1 配置 K3s 镜像仓库
启动 K3s 默认会从
docker.io
拉取镜像。使用
containerd
容器运行时在离线安装时,我们只需要将镜像仓库地址配置到
docker.io
下的
endpoint
即可,更多配置说明请参考配置
containerd
镜像仓库完全攻略或K3s 官方文档:
https://docs.rancher.cn/docs/k3s/installation/private-registry/_index/
sudomkdir -p /etc/rancher/k3s
sudocat>> /etc/rancher/k3s/registries.yaml <<EOF
mirrors:
"docker.io":
endpoint:
- "http://192.168.64.44:5000"
- "https://registry-1.docker.io"
EOF
4.2 授予可执行权限
sudochmod a+x /root/k3s /root/k3s-install.sh
sudocp /root/k3s /usr/local/bin/
4.3 安装 K3s
INSTALL_K3S_SKIP_DOWNLOAD=true /root/k3s-install.sh
稍等片刻,即可查看到 K3s 已经成功启动:
$ crictl images
IMAGE TAG IMAGE ID SIZE
docker.io/rancher/coredns-coredns 1.8.0 296a6d5035e2d 12.9MB
docker.io/rancher/klipper-helm v0.3.2 4be09ab862d40 50.7MB
docker.io/rancher/klipper-lb v0.1.2 897ce3c5fc8ff 2.71MB
docker.io/rancher/library-traefik 1.7.19 aa764f7db3051 24MB
docker.io/rancher/local-path-provisioner v0.0.14 e422121c9c5f9 13.4MB
docker.io/rancher/metrics-server v0.3.6 9dd718864ce61 10.5MB
docker.io/rancher/pause 3.1 da86e6ba6ca19 326kB
$ kubectl get pods -A
NAMESPACE NAME READY STATUS RESTARTS AGE
kube-system local-path-provisioner-7c458769fb-7w8hb 1/1 Running 0 37s
kube-system coredns-854c77959c-f8m2n 1/1 Running 0 37s
kube-system helm-install-traefik-9lbrx 0/1 Completed 0 38s
kube-system svclb-traefik-x8f6f 2/2 Running 0 29s
kube-system metrics-server-86cbb8457f-f7lb7 1/1 Running 0 37s
kube-system traefik-6f9cbd9bd4-4s66r 1/1 Running 0 29s
5. Docker + 私有镜像仓库方式
- 私有镜像仓库搭建教程请参考
设你已经将同一版本的 K3s 的安装脚本(k3s-install.sh)、K3s 的二进制文件(k3s)下载到了/root目录下。并且 K3s 所需要的镜像已经上传到了镜像仓库(本例的镜像仓库地址为:
http://192.168.64.44:5000
)。K3s 所需的镜像列表可以从 K3s Release页面的k3s-images.txt获得。
5.1 配置 K3s 镜像仓库
Docker 不支持像
containerd
那样可以通过修改
docker.io
对应的
endpoint
(默认为
https://registry-1.docker.io
)来间接修改默认镜像仓库的地址。但在Docker中可以通过配置registry-mirrors来实现从其他镜像仓库中获取K3s镜像。这样配置之后,会先从registry-mirrors配置的地址拉取镜像,如果获取不到才会从默认的docker.io获取镜像,从而满足了我们的需求。
cat>> /etc/docker/daemon.json <<EOF
{
"registry-mirrors": ["http://192.168.64.44:5000"]
}
EOFsudo systemctl daemon-reload
sudo systemctl restart docker
5.2 授予可执行权限
sudochmod a+x /root/k3s /root/k3s-install.sh
sudocp /root/k3s /usr/local/bin/
5.3 安装 K3s
INSTALL_K3S_SKIP_DOWNLOAD=true INSTALL_K3S_EXEC='--docker' /root/k3s-install.sh
稍等片刻,即可查看到 K3s 已经成功启动:
$ docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
rancher/klipper-helm v0.3.2 4be09ab862d4 7 weeks ago 145MB
rancher/coredns-coredns 1.8.0 296a6d5035e2 2 months ago 42.5MB
rancher/local-path-provisioner v0.0.14 e422121c9c5f 7 months ago 41.7MB
rancher/library-traefik 1.7.19 aa764f7db305 14 months ago 85.7MB
rancher/metrics-server v0.3.6 9dd718864ce6 14 months ago 39.9MB
rancher/klipper-lb v0.1.2 897ce3c5fc8f 19 months ago 6.1MB
rancher/pause 3.1 da86e6ba6ca1 3 years ago 742kB
$ kubectl get pods -A
NAMESPACE NAME READY STATUS RESTARTS AGE
kube-system helm-install-traefik-bcclh 0/1 Completed 0 33s
kube-system coredns-854c77959c-kp85f 1/1 Running 0 33s
kube-system metrics-server-86cbb8457f-85fpd 1/1 Running 0 33s
kube-system local-path-provisioner-7c458769fb-r5nkw 1/1 Running 0 33s
kube-system svclb-traefik-rbmhk 2/2 Running 0 24s
kube-system traefik-6f9cbd9bd4-k6t9n 1/1 Running 0 24s
6. 卸载
如果您使用安装脚本安装了 K3s,那么在安装过程中会生成一个卸载 K3s 的脚本。
卸载 K3s 会删除集群数据和所有脚本。要使用不同的安装选项重新启动集群,请使用不同的标志重新运行安装脚本。
要从 server 节点卸载 K3s,请运行:
/usr/local/bin/k3s-uninstall.sh
要从 agent 节点卸载 K3s,请运行:
/usr/local/bin/k3s-agent-uninstall.sh
参考:
- k3s 离线安装文档
- 一文搞定全场景K3s离线安装
- https://get.k3s.io/
- https://github.com/k3s-io/k3s/releases
版权归原作者 ghostwritten 所有, 如有侵权,请联系我们删除。