kubeasz/docs/setup/05-install_kube_node.md

119 lines
4.3 KiB
Markdown
Raw Normal View History

## 05-安装kube_node节点
2017-12-06 11:17:27 +08:00
`kube_node` 是集群中运行工作负载的节点,前置条件需要先部署好`kube_master`节点,它需要部署如下组件:
2017-12-06 11:17:27 +08:00
2022-06-28 19:59:50 +08:00
``` bash
cat playbooks/05.kube-node.yml
- hosts: kube_node
roles:
- { role: kube-lb, when: "inventory_hostname not in groups['kube_master']" }
- { role: kube-node, when: "inventory_hostname not in groups['kube_master']" }
```
+ kube-lb由nginx裁剪编译的四层负载均衡用于将请求转发到主节点的 apiserver服务
+ kubeletkube_node上最主要的组件
+ kube-proxy 发布应用服务与负载均衡
2017-12-06 11:17:27 +08:00
2018-01-05 23:05:22 +08:00
### 创建cni 基础网络插件配置文件
因为后续需要用 `DaemonSet Pod`方式运行k8s网络插件所以kubelet.server服务必须开启cni相关参数并且提供cni网络配置文件
2017-12-06 11:17:27 +08:00
### 创建 kubelet 的服务文件
+ 根据官方建议独立使用 kubelet 配置文件详见roles/kube-node/templates/kubelet-config.yaml.j2
2017-12-06 11:17:27 +08:00
+ 必须先创建工作目录 `/var/lib/kubelet`
``` bash
[Unit]
Description=Kubernetes Kubelet
Documentation=https://github.com/GoogleCloudPlatform/kubernetes
[Service]
WorkingDirectory=/var/lib/kubelet
2022-06-28 19:59:50 +08:00
{% if ansible_distribution == "Debian" and ansible_distribution_version|int >= 10 %}
ExecStartPre=/bin/mount -o remount,rw '/sys/fs/cgroup'
2022-06-28 19:59:50 +08:00
{% endif %}
{% if KUBE_RESERVED_ENABLED == "yes" or SYS_RESERVED_ENABLED == "yes" %}
ExecStartPre=/bin/mkdir -p /sys/fs/cgroup/cpu/podruntime.slice
ExecStartPre=/bin/mkdir -p /sys/fs/cgroup/cpuacct/podruntime.slice
ExecStartPre=/bin/mkdir -p /sys/fs/cgroup/cpuset/podruntime.slice
ExecStartPre=/bin/mkdir -p /sys/fs/cgroup/memory/podruntime.slice
ExecStartPre=/bin/mkdir -p /sys/fs/cgroup/pids/podruntime.slice
ExecStartPre=/bin/mkdir -p /sys/fs/cgroup/systemd/podruntime.slice
ExecStartPre=/bin/mkdir -p /sys/fs/cgroup/cpu/system.slice
ExecStartPre=/bin/mkdir -p /sys/fs/cgroup/cpuacct/system.slice
ExecStartPre=/bin/mkdir -p /sys/fs/cgroup/cpuset/system.slice
ExecStartPre=/bin/mkdir -p /sys/fs/cgroup/memory/system.slice
ExecStartPre=/bin/mkdir -p /sys/fs/cgroup/pids/system.slice
ExecStartPre=/bin/mkdir -p /sys/fs/cgroup/systemd/system.slice
{% if ansible_distribution != "Debian" %}
ExecStartPre=/bin/mkdir -p /sys/fs/cgroup/hugetlb/podruntime.slice
ExecStartPre=/bin/mkdir -p /sys/fs/cgroup/hugetlb/system.slice
{% endif %}
{% endif %}
2017-12-06 11:17:27 +08:00
ExecStart={{ bin_dir }}/kubelet \
--config=/var/lib/kubelet/config.yaml \
--container-runtime-endpoint=unix:///run/containerd/containerd.sock \
2018-09-17 22:20:52 +08:00
--hostname-override={{ inventory_hostname }} \
--kubeconfig=/etc/kubernetes/kubelet.kubeconfig \
--root-dir={{ KUBELET_ROOT_DIR }} \
2017-12-06 11:17:27 +08:00
--v=2
Restart=always
2017-12-06 11:17:27 +08:00
RestartSec=5
[Install]
WantedBy=multi-user.target
```
2019-04-10 23:49:49 +08:00
+ --ExecStartPre=/bin/mkdir -p xxx 对于某些系统centos7cpuset和hugetlb 是默认没有初始化system.slice 的,需要手动创建,否则在启用--kube-reserved-cgroup 时会报错Failed to start ContainerManager Failed to enforce System Reserved Cgroup Limits
+ 关于kubelet资源预留相关配置请参考 https://kubernetes.io/docs/tasks/administer-cluster/reserve-compute-resources/
2017-12-06 11:17:27 +08:00
### 创建 kube-proxy kubeconfig 文件
该步骤已经在 deploy节点完成[roles/deploy/tasks/main.yml](../../roles/deploy/tasks/main.yml)
2018-03-18 18:25:38 +08:00
2017-12-06 11:17:27 +08:00
+ 生成的kube-proxy.kubeconfig 配置文件需要移动到/etc/kubernetes/目录后续kube-proxy服务启动参数里面需要指定
### 创建 kube-proxy服务文件
``` bash
[Unit]
Description=Kubernetes Kube-Proxy Server
Documentation=https://github.com/GoogleCloudPlatform/kubernetes
After=network.target
[Service]
WorkingDirectory=/var/lib/kube-proxy
ExecStart={{ bin_dir }}/kube-proxy \
2022-06-28 19:59:50 +08:00
--config=/var/lib/kube-proxy/kube-proxy-config.yaml
Restart=always
2017-12-06 11:17:27 +08:00
RestartSec=5
LimitNOFILE=65536
[Install]
WantedBy=multi-user.target
```
2022-06-28 19:59:50 +08:00
请注意 [kube-proxy-config](../../roles/kube-node/templates/kube-proxy-config.yaml.j2) 文件的注释说明
2017-12-06 11:17:27 +08:00
2017-12-07 23:02:58 +08:00
### 验证 node 状态
``` bash
systemctl status kubelet # 查看状态
systemctl status kube-proxy
journalctl -u kubelet # 查看日志
journalctl -u kube-proxy
```
运行 `kubectl get node` 可以看到类似
``` bash
NAME STATUS ROLES AGE VERSION
2018-01-05 23:05:22 +08:00
192.168.1.42 Ready <none> 2d v1.9.0
192.168.1.43 Ready <none> 2d v1.9.0
192.168.1.44 Ready <none> 2d v1.9.0
2017-12-07 23:02:58 +08:00
```
2017-12-18 17:35:51 +08:00
2018-10-12 13:30:19 +08:00
[后一篇](06-install_network_plugin.md)