Kubernetes已经改变了如何大规模部署和管理容器化工作负载。现在开发人员面临的挑战主要在于设置过程的复杂性和资源需求量巨大。如果你深受内存不足的困扰，想要部署轻量级Kubernetes集群来减少内存占用，那么你一定要考虑由Rancher Labs发布的轻量级Kubernetes发行版——k3s。它把安装Kubernetes所需的一切文件都打包进一个40MB大小的二进制文件中，仅需512MB的RAM即可运行。非常适用于资源有限的环境，如边缘计算场景、IoT等。

在实际场景中，为了获得开发和测试的动力，节省开销，用户希望能够以最少的资源利用率和较低的硬件规格来部署Kubernetes。而k3s正好满足了这一需求，它能够在任何512MB RAM以上的设备上运行集群，如IoT设备或ARM驱动的设备。

既然k3s仅需少量资源即可运行，那么这意味着一些Kubernetes的特性被移除了：

旧的、非默认的、alpha功能
大部分in-tree插件（云提供商和存储插件），将其用附加组件进行替换
用sqlite来代替etcd作为默认存储机制

5分钟之内使用k3s部署轻量K8s集群

在本文中，我将使用运行在Debian 10上的3个server，每个server有1GB的RAM和1vcpu。其中一个server作为master，其他两个作为worker节点。

$ openstack server list
+--------------------------------------+-------------------+---------+-----------------------------------+-----------+-----------+
| ID                                   | Name              | Status  | Networks                          | Image     | Flavor    |
+--------------------------------------+-------------------+---------+-----------------------------------+-----------+-----------+
| 4df6a6dc-26e8-4ae0-8b6e-2f97daec0ef3 | k3s-master        | ACTIVE  | private=10.10.1.159               | Debian-10 | m1.tiny   |
| 5ca13239-b745-4f62-ab11-0a27949c9b35 | k3s-node02        | ACTIVE  | private=10.10.1.142               | Debian-10 | m1.tiny   |
| a54997f2-4d94-4718-86ab-73609b328761 | k3s-node01        | ACTIVE  | private=10.10.1.126               | Debian-10 | m1.tiny   |
+--------------------------------------+-------------------+---------+-----------------------------------+-----------+-----------+

我将在每个服务器的/ etc / hosts文件中为服务器添加A record。

sudo tee -a /etc/hosts<<EOF
10.10.1.159 k3s-master
10.10.1.126 k3s-node01
10.10.1.142 k3s-node02
EOF

##在Master节点上安装k3s

运行k3s的方式有很多，最快的方式是通过提供的bash脚本进行安装，同时该脚本提供了一个便捷的方式来安装到systemd或openrc。

curl -sfL https://get.k3s.io | sh -

安装输出：

[INFO]  Finding latest release
[INFO]  Using v0.8.1 as release
[INFO]  Downloading hash https://github.com/rancher/k3s/releases/download/v0.8.1/sha256sum-amd64.txt
[INFO]  Downloading binary https://github.com/rancher/k3s/releases/download/v0.8.1/k3s
[INFO]  Verifying binary download
[INFO]  Installing k3s to /usr/local/bin/k3s
[INFO]  Creating /usr/local/bin/kubectl symlink to k3s
[INFO]  Creating /usr/local/bin/crictl symlink to k3s
[INFO]  Creating /usr/local/bin/ctr symlink to k3s
[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 service file /etc/systemd/system/k3s.service
[INFO]  systemd: Enabling k3s unit
Created symlink /etc/systemd/system/multi-user.target.wants/k3s.service → /etc/systemd/system/k3s.service.
[INFO]  systemd: Starting k3s

安装完成之后，服务会自动启动。

$ systemctl status k3s
● k3s.service - Lightweight Kubernetes
   Loaded: loaded (/etc/systemd/system/k3s.service; enabled; vendor preset: enabled)
   Active: active (running) since Tue 2019-09-17 19:20:00 UTC; 2min 24s ago
     Docs: https://k3s.io
  Process: 833 ExecStartPre=/sbin/modprobe br_netfilter (code=exited, status=0/SUCCESS)
  Process: 836 ExecStartPre=/sbin/modprobe overlay (code=exited, status=0/SUCCESS)
 Main PID: 837 (k3s-server)
    Tasks: 98
   Memory: 571.1M
   CGroup: /system.slice/k3s.service
           ├─ 837 /usr/local/bin/k3s server KillMode=process
           ├─ 851 containerd -c /var/lib/rancher/k3s/agent/etc/containerd/config.toml -a /run/k3s/containerd/containerd.sock --state /run/k3s/conta
           ├─1110 containerd-shim -namespace k8s.io -workdir /var/lib/rancher/k3s/agent/containerd/io.containerd.runtime.v1.linux/k8s.io/f6eeb59978
           ├─1127 /pause
           ├─1207 containerd-shim -namespace k8s.io -workdir /var/lib/rancher/k3s/agent/containerd/io.containerd.runtime.v1.linux/k8s.io/0baf0ca181
           ├─1225 /coredns -conf /etc/coredns/Corefile
           ├─1576 containerd-shim -namespace k8s.io -workdir /var/lib/rancher/k3s/agent/containerd/io.containerd.runtime.v1.linux/k8s.io/dcce4b7e17
           ├─1594 /pause
           ├─1599 containerd-shim -namespace k8s.io -workdir /var/lib/rancher/k3s/agent/containerd/io.containerd.runtime.v1.linux/k8s.io/50816ffba8
           ├─1617 /pause
           ├─1824 containerd-shim -namespace k8s.io -workdir /var/lib/rancher/k3s/agent/containerd/io.containerd.runtime.v1.linux/k8s.io/d0ff393609
           ├─1842 /bin/sh /usr/bin/entry
           ├─1882 containerd-shim -namespace k8s.io -workdir /var/lib/rancher/k3s/agent/containerd/io.containerd.runtime.v1.linux/k8s.io/046779175f
           ├─1899 /bin/sh /usr/bin/entry
           ├─1904 containerd-shim -namespace k8s.io -workdir /var/lib/rancher/k3s/agent/containerd/io.containerd.runtime.v1.linux/k8s.io/93f0fe2361
           └─1921 /traefik --configfile=/config/traefik.toml

Sep 17 19:20:34 deb10 k3s[837]: E0917 19:20:34.714229     837 daemon_controller.go:302] kube-system/svclb-traefik failed with : error storing statu
Sep 17 19:20:34 deb10 k3s[837]: E0917 19:20:34.719452     837 daemon_controller.go:302] kube-system/svclb-traefik failed with : error storing statu
Sep 17 19:20:34 deb10 k3s[837]: I0917 19:20:34.726816     837 reconciler.go:207] operationExecutor.VerifyControllerAttachedVolume started for volum
Sep 17 19:20:34 deb10 k3s[837]: I0917 19:20:34.726836     837 reconciler.go:207] operationExecutor.VerifyControllerAttachedVolume started for volum
Sep 17 19:20:34 deb10 k3s[837]: I0917 19:20:34.726857     837 reconciler.go:207] operationExecutor.VerifyControllerAttachedVolume started for volum
Sep 17 19:20:34 deb10 k3s[837]: I0917 19:20:34.726869     837 reconciler.go:207] operationExecutor.VerifyControllerAttachedVolume started for volum
Sep 17 19:20:35 deb10 k3s[837]: I0917 19:20:35.529102     837 reconciler.go:181] operationExecutor.UnmountVolume started for volume "helm-traefik-t
Sep 17 19:20:35 deb10 k3s[837]: I0917 19:20:35.542858     837 operation_generator.go:799] UnmountVolume.TearDown succeeded for volume "kubernetes.i
Sep 17 19:20:35 deb10 k3s[837]: I0917 19:20:35.629277     837 reconciler.go:285] Volume detached for volume "helm-traefik-token-kjwrl" (UniqueName:
Sep 17 19:20:36 deb10 k3s[837]: W0917 19:20:36.355273     837 pod_container_deletor.go:75] C

将kubeconfig文件写入/etc/rancher/k3s/k3s.yaml：

$ cat /etc/rancher/k3s/k3s.yaml
cat: /etc/rancher/k3s/k3s.yaml: Permission denied
debian@deb10:~$ sudo cat /etc/rancher/k3s/k3s.yaml
apiVersion: v1
clusters:
- cluster:
    certificate-authority-data: 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
    server: https://localhost:6443
  name: default
contexts:
- context:
    cluster: default
    user: default
  name: default
current-context: default
kind: Config
preferences: {}
users:
- name: default
  user:
    password: 2d99cae31c075743be704bb717ceaae8
    username: admin

其他已经安装的有：

kubectl
crictl
k3s-killall.sh
k3s-uninstall.sh

在Worker节点上安装k3s

要在Woker节点上安装k3s，我们应该将K3S_URL以及K3S_TOKEN或K3S_CLUSTER_SECRET环境变量一起传递。

K3S_TOKEN在第一个节点上的/ var / lib / rancher / k3s / server / node-token中创建。

$ sudo cat /var/lib/rancher/k3s/server/node-token
K1042e2f8e353b9409472c1e0cca8457abe184dc7be3f0805109e92c50c193ceb42::node:c83acbf89a7de7026d6f6928dc270028

所以为了在worker节点上安装Kubernetes，我将运行：

k3s_url="https://k3s-master:6443"
k3s_token="K1042e2f8e353b9409472c1e0cca8457abe184dc7be3f0805109e92c50c193ceb42::node:c83acbf89a7de7026d6f6928dc270028"
curl -sfL https://get.k3s.io | K3S_URL=${k3s_url} K3S_TOKEN=${k3s_token} sh -

安装输出：

[INFO]  Finding latest release
[INFO]  Using v0.8.1 as release
[INFO]  Downloading hash https://github.com/rancher/k3s/releases/download/v0.8.1/sha256sum-amd64.txt
[INFO]  Downloading binary https://github.com/rancher/k3s/releases/download/v0.8.1/k3s
[INFO]  Verifying binary download
[INFO]  Installing k3s to /usr/local/bin/k3s
[INFO]  Creating /usr/local/bin/kubectl symlink to k3s
[INFO]  Creating /usr/local/bin/crictl symlink to k3s
[INFO]  Creating /usr/local/bin/ctr symlink to k3s
[INFO]  Creating killall script /usr/local/bin/k3s-killall.sh
[INFO]  Creating uninstall script /usr/local/bin/k3s-agent-uninstall.sh
[INFO]  env: Creating environment file /etc/systemd/system/k3s-agent.service.env
[INFO]  systemd: Creating service file /etc/systemd/system/k3s-agent.service
[INFO]  systemd: Enabling k3s-agent unit
Created symlink /etc/systemd/system/multi-user.target.wants/k3s-agent.service → /etc/systemd/system/k3s-agent.service.
[INFO]  systemd: Starting k3s-agent

登录到其中一个master节点并检查集群状态：

$ sudo kubectl config get-clusters 
NAME
default

$ sudo kubectl cluster-info 
Kubernetes master is running at https://localhost:6443
CoreDNS is running at https://localhost:6443/api/v1/namespaces/kube-system/services/kube-dns:dns/proxy

$ sudo kubectl get  nodes
NAME         STATUS   ROLES    AGE     VERSION
k3s-master   Ready    master   14m     v1.14.6-k3s.1
k3s-node01   Ready    worker   3m11s   v1.14.6-k3s.1
k3s-node02   Ready    worker   3m58s   v1.14.6-k3s.1

$ sudo kubectl get namespaces
NAME              STATUS   AGE
default           Active   16m
kube-node-lease   Active   16m
kube-public       Active   16m
kube-system       Active   16m

$ sudo  kubectl get endpoints -n kube-system
NAME       ENDPOINTS                                  AGE
kube-dns   10.42.0.2:53,10.42.0.2:53,10.42.0.2:9153   14m
traefik    10.42.0.5:80,10.42.0.5:443                 14m

$ sudo kubectl get pods -n kube-system
NAME                         READY   STATUS      RESTARTS   AGE
coredns-b7464766c-q9frk      1/1     Running     0          15m
helm-install-traefik-8dhpk   0/1     Completed   0          15m
svclb-traefik-9c2j8          2/2     Running     0          4m49s
svclb-traefik-bf9zd          2/2     Running     0          4m2s
svclb-traefik-v2fpx          2/2     Running     0          14m
traefik-5c79b789c5-k589d     1/1     Running     0          14m

使用crictl命令来查看正在运行的容器

# Master
$ sudo crictl ps
CONTAINER ID        IMAGE               CREATED             STATE               NAME                ATTEMPT             POD ID
acfafb50852d3       18471c10e6e4b       16 minutes ago      Running             traefik             0                   bf8534452389f
fee5ac7e88f2e       4a065d8dfa588       16 minutes ago      Running             lb-port-443         0                   e7068ff7ab2f2
bbab5b07e5efb       4a065d8dfa588       16 minutes ago      Running             lb-port-80          0                   e7068ff7ab2f2
65c5d1333ea04       2ee68ed074c6e       16 minutes ago      Running             coredns             0                   435c51f4716fc

# Workers
$ sudo crictl ps
CONTAINER ID        IMAGE               CREATED             STATE               NAME                ATTEMPT             POD ID
7ad5c83d6466f       4a065d8dfa588       6 minutes ago       Running             lb-port-443         0                   bf8d9fe57c3f3
c1380eabc0b33       4a065d8dfa588       6 minutes ago       Running             lb-port-80          0                   bf8d9fe57c3f3

大功告成啦！如果你需要更高级的配置，请参阅k3s文档：

https://rancher.com/docs/k3s/latest/en/

创作场景

Just 5 分钟！使用 k3s 部署轻量 Kubernetes 集群快速教程