写在前面
- 分享一个 k8s 高可用集群部署的 Ansible 剧本
- 以及涉及到的一些工具的安装
- 博文内容涉及：
- 从零开始 一个 k8s 高可用 集群部署 Ansible剧本编写，
- 编写后搭建 k8s 高可用 集群
- 一些集群常用的 监控，备份工具安装，包括：
cadvisor监控工具部署metrics-server监控工具部署Ingress—nginxIngress 控制器部署Metallb软 LoadBalancer 部署local-path-storage基于本地存储的SC 分配器部署prometheus监控工具部署ETCD快照备份定时任务编写运行Velero集群容灾备份工具部署
- 理解不足小伙伴帮忙指正
我所渴求的，無非是將心中脫穎語出的本性付諸生活，為何竟如此艱難呢 ------赫尔曼·黑塞《德米安》
部署完的 集群 node 信息查看,感觉版本有点高，生产不太建议安装这么高的版本，有些开源工具都安装不了，各种问问题…
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$kubectl get nodes
NAME STATUS ROLES AGE VERSION
vms100.liruilongs.github.io Ready control-plane 10d v1.25.1
vms101.liruilongs.github.io Ready control-plane 10d v1.25.1
vms102.liruilongs.github.io Ready control-plane 10d v1.25.1
vms103.liruilongs.github.io Ready <none> 10d v1.25.1
vms105.liruilongs.github.io Ready <none> 10d v1.25.1
vms106.liruilongs.github.io Ready <none> 10d v1.25.1
vms107.liruilongs.github.io Ready <none> 10d v1.25.1
vms108.liruilongs.github.io Ready <none> 10d v1.25.1
部署涉及机器
- master1: 192.168.26.100
- master2: 192.168.26.101
- master3: 192.168.26.102
- work Node1: 192.168.26.103
- work Node2: 192.168.26.105
- work Node3: 192.168.26.106
- work Node4: 192.168.26.107
- work Node5: 192.168.26.108
哈 104 不吉利，跳过去了…
集群部署
部署拓扑
下图为 k8s 官网文档中 HA 拓扑图，这里使用下面的拓扑方式部署
堆叠(Stacked)HA 集群拓扑， 其中 etcd 分布式数据存储集群堆叠在 kubeadm 管理的控制平面节点上，作为控制平面的一个组件运行。
部署后的 etcd 分布
┌──[root@vms100.liruilongs.github.io]-[~/ansible/kubescape]
└─$ETCDCTL_API=3 etcdctl --endpoints https://127.0.0.1:2379 --cert="/etc/kubernetes/pki/etcd/server.crt" --key="/etc/kubernetes/pki/etcd/server.key" --cacert="/etc/kubernetes/pki/etcd/ca.crt" member list -
w table
+------------------+---------+-----------------------------+-----------------------------+-----------------------------+
| ID | STATUS | NAME | PEER ADDRS | CLIENT ADDRS |
+------------------+---------+-----------------------------+-----------------------------+-----------------------------+
| ee392e5273e89e2 | started | vms100.liruilongs.github.io | https://192.168.26.100:2380 | https://192.168.26.100:2379 |
| 11486647d7f3a17b | started | vms102.liruilongs.github.io | https://192.168.26.102:2380 | https://192.168.26.102:2379 |
| e00e3877df8f76f4 | started | vms101.liruilongs.github.io | https://192.168.26.101:2380 | https://192.168.26.101:2379 |
+------------------+---------+-----------------------------+-----------------------------+-----------------------------+
每个控制平面节点运行 kube-apiserver、kube-scheduler 和 kube-controller-manager 实例。 kube-apiserver 使用负载均衡器暴露给工作节点。
每个控制平面节点创建一个本地 etcd 成员(member)，这个 etcd 成员只与该节点的 kube-apiserver 通信。 这同样适用于本地 kube-controller-manager 和 kube-scheduler 实例。
这种拓扑将控制平面和 etcd 成员耦合在同一节点上。相对使用外部 etcd 集群， 设置起来更简单，而且更易于副本管理。
然而，堆叠集群存在耦合失败的风险。如果一个节点发生故障，则 etcd 成员和控制平面实例都将丢失， 并且冗余会受到影响。你可以通过添加更多控制平面节点来降低此风险。
因此，你应该为 HA 集群运行至少三个堆叠的控制平面节点。
┌──[root@vms100.liruilongs.github.io]-[~/ansible/kubescape]
└─$ETCDCTL_API=3 etcdctl --endpoints https://127.0.0.1:2379 --cert="/etc/kubernetes/pki/etcd/server.crt" --key="/etc/kubernetes/pki/etcd/server.key" --cacert="/etc/kubernetes/pki/etcd/ca.crt" endpoint status --cluster -w table
+-----------------------------+------------------+---------+---------+-----------+-----------+------------+
| ENDPOINT | ID | VERSION | DB SIZE | IS LEADER | RAFT TERM | RAFT INDEX |
+-----------------------------+------------------+---------+---------+-----------+-----------+------------+
| https://192.168.26.100:2379 | ee392e5273e89e2 | 3.5.4 | 37 MB | false | 100 | 3152364 |
| https://192.168.26.102:2379 | 11486647d7f3a17b | 3.5.4 | 36 MB | false | 100 | 3152364 |
| https://192.168.26.101:2379 | e00e3877df8f76f4 | 3.5.4 | 36 MB | true | 100 | 3152364 |
+-----------------------------+------------------+---------+---------+-----------+-----------+------------+
┌──[root@vms100.liruilongs.github.io]-[~/ansible/kubescape]
└─$
当使用 kubeadm init 和 kubeadm join --control-plane 时， 在控制平面节点上会自动创建本地 etcd 成员。ETCD 一点要定期备份
部署中涉及太多文件，篇幅问题，文件无法展示，以上传 git 仓库，获取 git 仓库地址方式，关注 公总好 山河已无恙，回复 k8s-ha-deploy 即可获得地址。
免密配置
这里假设我们拿到的是安装了系统的新机。这里我的机器，通过 克隆的的生成的 虚机，所以配置了 YUM 源，如果没有，需要先配置一台虚机，剩下的通过 Ansible 的批量配置下 YUM ，所以对应 YUM 的配置这个教程没有涉及，YUM 主要配置为 阿里云的 就可以。配置文件小伙伴可以通过 git 获取。
在安装 Ansible 之前，方便操作，我们需要批量配置下免密，这个使用 expect 的方式实现，用直接写好了 脚本直接执行，需要配置的主机 单独列出读取。
安装 expect
┌──[root@vms100.liruilongs.github.io]-[~]
└─$yum -y install expect
列出部署主机
┌──[root@vms100.liruilongs.github.io]-[~]
└─$cat host_list
192.168.26.100
192.168.26.101
192.168.26.102
192.168.26.103
192.168.26.105
192.168.26.106
192.168.26.107
192.168.26.108
免密脚本，直接执行即可，如果主机清单文件名相同，直接读取即可
#!/bin/bash
#@File : mianmi.sh
#@Time : 2022/08/20 17:45:53
#@Author : Li Ruilong
#@Version : 1.0
#@Desc : None
#@Contact : 1224965096@qq.com
/usr/bin/expect <<-EOF
spawn ssh-keygen
expect "(/root/.ssh/id_rsa)" {send "\r"}
expect {
"(empty for no passphrase)" {send "\r"}
"already" {send "y\r"}
}
expect {
"again" {send "\r"}
"(empty for no passphrase)" {send "\r"}
}
expect {
"again" {send "\r"}
"#" {send "\r"}
}
expect "#"
expect eof
EOF
for IP in $( cat host_list )
do
if [ -n IP ];then
/usr/bin/expect <<-EOF
spawn ssh-copy-id root@$IP
expect {
"*yes/no*" { send "yes\r"}
"*password*" { send "redhat\r" }
}
expect {
"*password" { send "redhat\r"}
"#" { send "\r"}
}
expect "#"
expect eof
EOF
fi
done
OK，配置完免密之后，需要安装 Ansible， 运维工具，一定要装一个，机器太多了很费人。
Ansible 安装
┌──[root@vms100.liruilongs.github.io]-[~]
└─$yum -y install ansible
ansible 配置
这里的主机清单，随便配置一个无效的，建议不要在配置文件中配置正确的主机清单，防止执行剧本到错误的主机节点。
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$vim ansible.cfg
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$cat ansible.cfg
[defaults]
# 主机清单文件，就是要控制的主机列表
inventory=inventory
# 连接受管机器的远程的用户名
remote_user=root
# 角色目录
roles_path=roles
# 设置用户的su 提权
[privilege_escalation]
become=True
become_method=sudo
become_user=root
become_ask_pass=False
拷贝前面的文件测试下网络情况
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$cat host_list
192.168.26.100
192.168.26.101
192.168.26.102
192.168.26.103
192.168.26.105
192.168.26.106
192.168.26.107
192.168.26.108
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$ansible all -m ping -i host_list
测试没问题，需要编写下 主机清单文件
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$cat host.yaml
ansible:
children:
ansible_master:
hosts:
192.168.26.100:
ansible_node:
hosts:
192.168.26.[101:103]:
192.168.26.[105:108]:
k8s:
children:
k8s_master:
hosts:
192.168.26.[100:102]:
k8s_node:
hosts:
192.168.26.103:
192.168.26.[105:108]:
检查清单文件，这里的分组就不多讲了。
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$ansible-inventory -i host.yaml --graph
@all:
|--@ansible:
| |--@ansible_master:
| | |--192.168.26.100
| |--@ansible_node:
| | |--192.168.26.101
| | |--192.168.26.102
| | |--192.168.26.103
| | |--192.168.26.105
| | |--192.168.26.106
| | |--192.168.26.107
| | |--192.168.26.108
|--@k8s:
| |--@k8s_master:
| | |--192.168.26.100
| | |--192.168.26.101
| | |--192.168.26.102
| |--@k8s_node:
| | |--192.168.26.103
| | |--192.168.26.105
| | |--192.168.26.106
| | |--192.168.26.107
| | |--192.168.26.108
|--@ungrouped:
测试一下清单文件
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$ansible all --list-hosts -i host.yaml
hosts (8):
192.168.26.100
192.168.26.101
192.168.26.102
192.168.26.103
192.168.26.105
192.168.26.106
192.168.26.107
192.168.26.108
为了使用 tab 键，这里我们建一些清单组的空文件。
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$touch ansible_master ansible_node k8s_master k8s_node
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$ls
ansible.cfg ansible_node host_list_no_ansible k8s_master ps1_mod.yaml
ansible_master host_list host.yaml k8s_node
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$
为了方便演示，看到当前的执行目录，配置下 PS1
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$cat ps1_mod.yaml
---
- name: all modify PS1
hosts: ansible_node
tasks:
- name: PS1 modify
shell: echo 'PS1="\[\033[1;32m\]┌──[\[\033[1;34m\]\u@\H\[\033[1;32m\]]-[\[\033[0;1m\]\w\[\033[1;32m\]] \n\[\033[1;32m\]└─\[\033[1;34m\]\$\[\033[0m\]"' >> /root/.bashrc
- name: PS1 disply
shell: cat /root/.bashrc | grep PS1
执行剧本并测试。
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$ansible-playbook ps1_mod.yaml -i host.yaml -vv
k8s 安装前环境准备
检查机器 UUID 和 网卡MAC 地址
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$ansible k8s -m shell -a "ip link | grep ether| awk '{print $2 }' " -i host.yaml
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$ansible k8s -m shell -a "cat /sys/class/dmi/id/product_uuid " -i host.yaml
剩下的 操作这个编写一个 k8s 部署环境的初始化的 剧本任务 k8s_init_deploy.yaml,涉及各项通过剧本任务引用实现
hosts 文件准备
没有 DNS ，如果有 DNS 服务器，则不需要
┌──[root@vms100.liruilongs.github.io]-[~/ansible/file]
└─$cat hosts
127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4
::1 localhost localhost.localdomain localhost6 localhost6.localdomain6
192.168.26.100 vms100.liruilongs.github.io vms100
192.168.26.101 vms101.liruilongs.github.io vms101
192.168.26.102 vms102.liruilongs.github.io vms102
192.168.26.103 vms103.liruilongs.github.io vms103
192.168.26.105 vms105.liruilongs.github.io vms105
192.168.26.106 vms106.liruilongs.github.io vms106
192.168.26.107 vms107.liruilongs.github.io vms107
192.168.26.108 vms108.liruilongs.github.io vms108
把 hosts 文件替换为所有机器的 hosts ，编写剧本任务
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$cat k8s_init_deploy.yaml
---
- name: copy "/etc/hosts"
copy:
src: ./file/hosts
dest: /etc/hosts
force: yes
防火墙，交换分区，SElinux 设置
编写剧本任务，在之前的 初始化剧本补充。
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$cat k8s_init_deploy.yaml
---
# 拷贝 hosts 文件,没有 DNS 需要
- name: copy "/etc/hosts"
copy:
src: ./file/hosts
dest: /etc/hosts
force: yes
# 关闭防火墙,这里设置为 trusted ,以后可能处理漏洞使用
- name: firewalld setting trusted
firewalld:
zone: trusted
permanent: yes
state: enabled
# 关闭 SELinux
- name: Disable SELinux
selinux:
state: disabled
# 禁用交换分区
- name: Disable swapoff
shell: /usr/sbin/swapoff -a
# 删除 swap 配置
- name: delete /etc/fstab
shell: sed -i '/swap/d' /etc/fstab
这里我们通过 任务引用的方式来执行，并且打一个 init 的标签。方便之后单独执行，引入到当前的 k8s 部署剧本 k8s_deploy.yaml
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$cat k8s_deploy.yaml
---
- name: k8s deploy
hosts: k8s
tasks:
- name: init k8s
include_tasks:
file: k8s_init_deploy.yaml
tags:
- init_k8s
k8s_deploy.yaml 为部署 k8s 所有操作的剧本。可以先执行下，完成前面的部署工作。
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$ansible-playbook k8s_deploy.yaml -i host.yaml -f 4
有 8 个 机器，这里的 -f 4 为并行执行的意思。 -i host.yaml 指定主机清单
安装容器运行时 CRI docker
这里选择 Dokcer 作为 CRI ，但是Docker 本身没有实现 CRI，在 k8s 在 1.24 的移除了 docker 和 K8s 的桥梁 Dockershim ，所以不能直接使用，需要安装 cri-docker .
这里需要注意：
部署 cri-docker 要重载沙箱(pause)镜像，cri-dockerd 适配器能够接受指定用作 Pod 的基础容器的容器镜像(“pause 镜像”)作为命令行参数。 要使用的命令行参数是 --pod-infra-container-image。 如果不指定，会直接从 谷歌的镜像库拉取，即使在 kubeadm init 指定了 镜像库也不行(我部署是不行)，同时，如果部署 HA ，使用 keepalived + Haproxy 的方式，并且通过 静态 Pod 的方式部署，在 启动 kubelet 的时候 静态pod 也会直接从 谷歌的镜像库拉取。
对应 cgroup 驱动的使用，按照官方官网推荐的来。 如果系统通过 systemd 来引导，就用 systemd
CRI 安装的初始化工作
转发 IPv4 并让 iptables 看到桥接流量, 相关文件准备，需要注意
┌──[root@vms100.liruilongs.github.io]-[~/ansible/file/modules-load.d]
└─$cat k8s.conf
overlay
br_netfilter
┌──[root@vms100.liruilongs.github.io]-[~/ansible/file/sysctl.d]
└─$cat k8s.conf
net.bridge.bridge-nf-call-iptables = 1
net.bridge.bridge-nf-call-ip6tables = 1
net.ipv4.ip_forward = 1
docker 配置文件准备，这里的配置项 "data-root": "/docker/data", 为 docker 数据目录，最好找一个大一点的地方。如果修改的话，剧本的创建目录要同步修改。
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$cat file/daemon.json
{
"exec-opts": ["native.cgroupdriver=systemd"],
"log-driver": "json-file",
"data-root": "/docker/data",
"log-opts": {
"max-size": "100m"
},
"storage-driver": "overlay2",
"storage-opts": [
"overlay2.override_kernel_check=true"
],
"experimental": false,
"debug": false,
"max-concurrent-downloads": 10,
"registry-mirrors": [
"https://2tefyfv7.mirror.aliyuncs.com",
"https://docker.mirrors.ustc.edu.cn"
]
}
当前需要复制的配置文件
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$ls file/*
file/daemon.json file/hosts
file/modules-load.d:
k8s.conf
file/sysctl.d:
k8s.conf
cri-dockerd rpm 包下载，下载到 Anasible 控制节点，复制到其他的节点。
PS C:\Users\山河已无恙\Downloads> curl -o cri-dockerd-0.3.0-3.el7.x86_64.rpm https://github.com/Mirantis/cri-dockerd/releases/download/v0.3.0/cri-dockerd-0.3.0-3.el7.x86_64.rpm
PS C:\Users\山河已无恙\Downloads> scp .\cri-dockerd-0.3.0-3.el7.x86_64.rpm root@192.168.26.100:/root/ansible/install_package
root@192.168.26.100's password:
cri-dockerd-0.3.0-3.el7.x86_64.rpm 100% 9195KB 81.7MB/s 00:00
PS C:\Users\山河已无恙\Downloads>
存放位置
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$cd install_package/;ls
cri-dockerd-0.3.0-3.el7.x86_64.rpm
CRI docker, 安装 任务剧本
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$cat k8s_cri_deploy.yaml
# ansible 2.9.27
# #@File : k8s_cri_deploy.yaml
# #@Time : 2023/01/19 22:32:47
# #@Author : Li Ruilong
# #@Version : 1.0
# #@Desc : 安装 CRI , 这里我们选择 Docker,需要安装 cri-docker
# #@Contact : liruilonger@gmail.com
---
# 转发 IPv4 并让 iptables 看到桥接流量
- name: Forwarding IPv4 and letting iptables see bridged traffic 1
copy:
src: ./file/modules-load.d/k8s.conf
dest: /etc/modules-load.d/k8s.conf
- name: Forwarding IPv4 and letting iptables see bridged traffic 2
shell: modprobe overlay && modprobe br_netfilter
# 永久设置所需的 sysctl 参数
- name: sysctl params required by setup, params persist across reboots
copy:
src: ./file/sysctl.d/k8s.conf
dest: /etc/sysctl.d/k8s.conf
# 刷新内核参数
- name: Apply sysctl params without reboot
shell: sysctl --system
# 安装 docker
- name: install docker-ce docker-ce-cli containerd.io
yum:
name:
- docker-ce
- docker-ce-cli
- containerd.io
state: present
# 创建 dockers 数据目录
- name: create docker data dir
file:
path: /docker/data
state: directory
# 创建 docker 配置目录
- name: create docker data dir
file:
path: /etc/docker
state: directory
# 修改 dockers 配置
- name: modify docker config
copy:
src: ./file/daemon.json
dest: /etc/docker/daemon.json
# 复制 cri-docker rpm 安装包
- name: copy install cri-docker rpm
# https://github.com/Mirantis/cri-dockerd/releases/download/v0.3.0/cri-dockerd-0.3.0-3.el7.x86_64.rpm
copy:
src: ./install_package/cri-dockerd-0.3.0-3.el7.x86_64.rpm
dest: /tmp/cri-dockerd-0.3.0-3.el7.x86_64.rpm
# 安装 cri-docker
- name: install cri-docker
yum:
name: /tmp/cri-dockerd-0.3.0-3.el7.x86_64.rpm
state: present
# 修改 cri-docker service 文件，否则 沙箱会从谷歌的仓库拉去
- name: modify cri-dockerd service file
copy:
src: ./file/cri-docker.service
dest: /usr/lib/systemd/system/cri-docker.service
# 配置开机自启,启动 docker
- name: start docker, setting enable
service:
name: docker
state: restarted
enabled: yes
# 配置开机自启,启动 cri-docker
- name: start cri-docker, setting enable
service:
name: cri-docker
state: restarted
enabled: yes
# 配置开机自启,启动 cri-docker.socket
- name: start cri-docker,socket setting enable
service:
name: cri-docker.socket
enabled: yes
# 配置校验
- name: check init cri
include_tasks:
file: k8s_cri_deploy_check.yaml
tags: cri_check
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$
安装完的校验单独放到了 k8s_cri_deploy_check.yaml 位置，也是通过剧本引用的方式，并且打了 cri_check 标签。这个如果希望看到检查数据，需要执行剧本时加上 -vv,这里只是写了几个基本的校验，还可以对其他的做校验
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$cat k8s_cri_deploy_check.yaml
# ansible 2.9.27
# #@File : k8s_cri_deploy_check.yaml
# #@Time : 2023/01/19 23:02:47
# #@Author : Li Ruilong
# #@Version : 1.0
# #@Desc : cri(docker) 配置检查剧本任务
# #@Contact : liruilonger@gmail.com
---
- name: check br_netfilter, overlay
shell: (lsmod | grep br_netfilter ;lsmod | grep overlay)
- name: charck ipv4 sysctl param
shell: sysctl net.bridge.bridge-nf-call-iptables net.bridge.bridge-nf-call-ip6tables net.ipv4.ip_forward
- name: chrck docker config
shell: docker info
当前的 k8s 安装剧本，引入了 k8s_cri_deploy.yaml 的安装，并且 打标签 cri。
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$cat k8s_deploy.yaml
# ansible 2.9.27
# #@File : k8s_deploy.yaml
# #@Time : 2023/01/19 23:02:47
# #@Author : Li Ruilong
# #@Version : 1.0
# #@Desc : k8s 安装剧本
# #@Contact : liruilonger@gmail.com
---
- name: k8s deploy 1
hosts: k8s
tasks:
# 初始化 K8s 安装环境
- name: init k8s
include_tasks:
file: k8s_init_deploy.yaml
tags:
- init_k8s
# 安装 CRI docker ,cri-docker
- name: CRI deploy (docker,cri-docker)
include_tasks:
file: k8s_cri_deploy.yaml
tags: cri
可以执行测试下
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$ansible-playbook k8s_deploy.yaml -i host.yaml -f 4
安装 kubeadm、kubelet 和 kubectl
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$cat k8s_kubeadm_kubelet_kubectl_deploy.yaml
# ansible 2.9.27
# #@File : k8s_kubeadm_kubelet_kubectl_deploy.yaml
# #@Time : 2023/01/19 23:30:47
# #@Author : Li Ruilong
# #@Version : 1.0
# #@Desc : 安装 kubeadm、kubelet 和 kubectl
# #@Contact : liruilonger@gmail.com
---
# 安装 kubeadm、kubelet 和 kubectl
- name: install kubeadm、kubelet and kubectl
yum:
name:
- kubelet-1.25.1-0
- kubeadm-1.25.1-0
- kubectl-1.25.1-0
state: present
# 启动 kubelet 并设置开启自启
- name: start kubelet,setting enable
service:
name: kubelet
state: started
enabled: yes
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$
添加的部署脚本
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$cat k8s_deploy.yaml
# ansible 2.9.27
# #@File : k8s_deploy.yaml
# #@Time : 2023/01/19 23:02:47
# #@Author : Li Ruilong
# #@Version : 1.0
# #@Desc : k8s 安装剧本
# #@Contact : liruilonger@gmail.com
---
- name: k8s deploy 1
hosts: k8s
tasks:
# 初始化 K8s 安装环境
- name: init k8s
include_tasks:
file: k8s_init_deploy.yaml
tags:
- init_k8s
# 安装 CRI docker ,cri-docker
- name: CRI deploy (docker,cri-docker)
include_tasks:
file: k8s_cri_deploy.yaml
tags: cri
# 安装 kubeadm,kubelet,kubectl
- name: install kubeadm,kubelet,kubectl
include_tasks:
file: k8s_kubeadm_kubelet_kubectl_deploy.yaml
tags: install_k8s
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$
HA 涉及的 keepalived、HAproxy 静态 Pod 准备
HA 涉及的 keepalived、HAproxy 通过 静态Pod 的方式运行，当前集群设置三个master 节点，kubelet 和 api-service 交互通过 keepalived 提供的 VIP 访问。然后由 HAprxy 把VIP地址反向代理到各 master 的 api-service 服务
涉及的到文件太多，这个不做展示
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$tree -h file/
file/
├── [ 25] haproxy
│ └── [1.8K] haproxy.cfg
├── [ 55] keepalived
│ ├── [ 370] check_apiserver.sh
│ └── [ 521] keepalived.conf
└── [ 49] manifests
├── [ 543] haproxy.yaml
└── [ 676] keepalived.yaml
对应的 任务剧本 manifests_keepalived_haproxy.yaml
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$cat manifests_keepalived_haproxy.yaml
# ansible 2.9.27
# # #@File : manifests_keepalived_haproxy.yaml
# # #@Time : 2023/01/19 23:02:47
# # #@Author : Li Ruilong
# # #@Version : 1.0
# # #@Desc : HA 静态 pod 相关 配置文件 YAML 文件准备
# # #@Contact : liruilonger@gmail.com
---
# 创建 静态 pod 目录
- name: create manifests dir
file:
path: /etc/kubernetes/manifests/
state: directory
force: true
# 复制 keepalived haproxy 对应的 静态 pod yaml 文件
- name: copy manifests pod, haproxy and keepalived
copy:
src: ./file/manifests/keepalived.yaml
dest: /etc/kubernetes/manifests/keepalived.yaml
- name: copy manifests pod, haproxy and keepalived
copy:
src: ./file/manifests/haproxy.yaml
dest: /etc/kubernetes/manifests/haproxy.yaml
# 创建 haproxy 配置文件 目录
- name: create haproxy dir
file:
path: /etc/haproxy
state: directory
force: true
- name: copy /etc/haproxy/haproxy.cfg
copy:
src: ./file/haproxy/haproxy.cfg
dest: /etc/haproxy/haproxy.cfg
# 创建 keepalived 配置文件 目录
- name: create keepalived dir
file:
path: /etc/keepalived
state: directory
force: true
- name: copy /etc/keepalived/keepalived.conf
copy:
src: ./file/keepalived/keepalived.conf
dest: /etc/keepalived/keepalived.conf
- name: copy /etc/keepalived/check_apiserver.sh
copy:
src: ./file/keepalived/check_apiserver.sh
dest: /etc/keepalived/check_apiserver.sh
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$
涉及到 HAproxy 和 keepalived 的 配置文件以及静态pod yaml文件的定义，篇幅原因，这里不做展示，可以参考官方文档给出的，如果实在不想找， 也可以通过 git 地址获取。
这个时候的部署剧本
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$cat k8s_deploy.yaml
# ansible 2.9.27
# #@File : k8s_deploy.yaml
# #@Time : 2023/01/19 23:02:47
# #@Author : Li Ruilong
# #@Version : 1.0
# #@Desc : k8s 安装剧本
# #@Contact : liruilonger@gmail.com
---
- name: k8s deploy 1
hosts: k8s
tasks:
# 初始化 K8s 安装环境
- name: init k8s
include_tasks:
file: k8s_init_deploy.yaml
tags:
- init_k8s
# 安装 CRI docker ,cri-docker
- name: CRI deploy (docker,cri-docker)
include_tasks:
file: k8s_cri_deploy.yaml
tags: cri
# 安装 kubeadm,kubelet,kubectl
- name: install kubeadm,kubelet,kubectl
include_tasks:
file: k8s_kubeadm_kubelet_kubectl_deploy.yaml
tags: install_k8s
- name: k8s deploy 2
hosts: k8s_master
tasks:
- name: manifests_keepalived_haproxy.yaml
include_tasks:
file: manifests_keepalived_haproxy.yaml
tags: install_k8s
初始化集群添加的控制节点，工作节点
先在一个 maste 节点执行，剩下的 master 节点通过添加的加入
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$kubeadm init --image-repository "registry.aliyuncs.com/google_containers" --control-plane-endpoint "192.168.26.99:30033" --upload-certs --kubernetes-version=v1.25.1 --pod-network-cidr=10.244.0.0/16 --cri-socket /var/run/cri-dockerd.sock
W0121 02:49:21.251663 106843 initconfiguration.go:119] Usage of CRI endpoints without URL scheme is deprecated and can cause kubelet errors in the future. Automatically prepending scheme "unix" to the "criSocket" with value "/var/run/cri-dockerd.sock". Please update your configuration!
[init] Using Kubernetes version: v1.25.1
[preflight] Running pre-flight checks
[WARNING Firewalld]: firewalld is active, please ensure ports [6443 10250] are open or your cluster may not function correctly
[preflight] Pulling images required for setting up a Kubernetes cluster
......................
[addons] Applied essential addon: CoreDNS
W0121 02:49:31.282997 106843 endpoint.go:57] [endpoint] WARNING: port specified in controlPlaneEndpoint overrides bindPort in the controlplane address
[addons] Applied essential addon: kube-proxy
Your Kubernetes control-plane has initialized successfully!
To start using your cluster, you need to run the following as a regular user:
mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config
Alternatively, if you are the root user, you can run:
export KUBECONFIG=/etc/kubernetes/admin.conf
You should now deploy a pod network to the cluster.
Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at:
https://kubernetes.io/docs/concepts/cluster-administration/addons/
You can now join any number of the control-plane node running the following command on each as root:
kubeadm join 192.168.26.99:30033 --token hdse4x.7d8vdil1s7tdp8dj \
--discovery-token-ca-cert-hash sha256:11bf0456c84c2a176eb380a98844eed80a296cdb6cc557c863991c63de594108 \
--control-plane --certificate-key 61b1a22aa5dbbd93660709aa343452d80ec98bf2efef6915b3a36ae97a8eefed
Please note that the certificate-key gives access to cluster sensitive data, keep it secret!
As a safeguard, uploaded-certs will be deleted in two hours; If necessary, you can use
"kubeadm init phase upload-certs --upload-certs" to reload certs afterward.
Then you can join any number of worker nodes by running the following on each as root:
kubeadm join 192.168.26.99:30033 --token hdse4x.7d8vdil1s7tdp8dj \
--discovery-token-ca-cert-hash sha256:11bf0456c84c2a176eb380a98844eed80a296cdb6cc557c863991c63de594108
第一个控制节点 准备 kubectl 客户端
mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config
CNI 插件准备
k8s 版本 和 CNI 的版本关系，以及安装相关可以通过下的地址查看。
https://projectcalico.docs.tigera.io/archive/v3.24/getting-started/kubernetes/self-managed-onprem/onpremises
https://projectcalico.docs.tigera.io/archive/v3.24/getting-started/kubernetes/requirements
这里需要注意的是:
好像 bpf 文件系统，需要高版本的 内核才支撑。我最开始的内核版本为 3.10 的版本，所有不支持，需要把 bpf 的挂载删掉。
我本地测试发现 calico v3.20 到 v3.25 都是需要的做处理的。 涉及的镜像最好导入进去，网络情况不同，拉取很费事。
┌──[root@vms100.liruilongs.github.io]-[~/ansible/calico]
└─$cat calico.v3.23.yaml | grep -A 3 -e bpffs$
- name: bpffs
mountPath: /sys/fs/bpf
- name: cni-log-dir
mountPath: /var/log/calico/cni
--
- name: bpffs
hostPath:
path: /sys/fs/bpf
type: Directory
需要删除的部分
。。。。。
- name: bpffs
mountPath: /sys/fs/bpf
。。。。。
- name: bpffs
hostPath:
path: /sys/fs/bpf
type: Directory
下载 YAML 文件部署
┌──[root@vms100.liruilongs.github.io]-[~/ansible/calico]
└─$wget --no-check-certificate https://docs.projectcalico.org/manifests/calico.yaml
--2023-01-21 02:56:01-- https://docs.projectcalico.org/manifests/calico.yaml
正在解析主机 docs.projectcalico.org (docs.projectcalico.org)... 18.139.194.139, 52.74.166.77, 2406:da18:880:3801::c8, ...
正在连接 docs.projectcalico.org (docs.projectcalico.org)|18.139.194.139|:443... 已连接。
警告: 无法验证 docs.projectcalico.org 的由 “/C=US/O=Let's Encrypt/CN=R3” 颁发的证书:
颁发的证书已经过期。
已发出 HTTP 请求，正在等待回应... 200 OK
长度：238089 (233K) [text/yaml]
正在保存至: “calico.yaml”
100%[=======================================================================================================================================================================>] 238,089 592KB/s 用时 0.4s
2023-01-21 02:56:02 (592 KB/s) - 已保存 “calico.yaml” [238089/238089])
┌──[root@vms100.liruilongs.github.io]-[~/ansible/calico]
└─$vim calico.yaml
修改 CALICO_IPV4POOL_CIDR 为之前指定的 地址
┌──[root@vms100.liruilongs.github.io]-[~/ansible/calico]
└─$cat calico.yaml | grep -C 3 IPV4POOL_CIDR
# The default IPv4 pool to create on startup if none exists. Pod IPs will be
# chosen from this range. Changing this value after installation will have
# no effect. This should fall within `--cluster-cidr`.
- name: CALICO_IPV4POOL_CIDR
value: "10.244.0.0/16"
# Disable file logging so `kubectl logs` works.
- name: CALICO_DISABLE_FILE_LOGGING
┌──[root@vms100.liruilongs.github.io]-[~/ansible/calico]
└─$
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$kubectl apply -f ./calico/calico.yaml
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$source <(kubectl completion bash) >> /etc/profile
其他控制节点添加
这个只有3个控制节点，所以通过 命令行单独处理，多的话可以使用 ansible
┌──[root@vms101.liruilongs.github.io]-[~]
└─$kubeadm join 192.168.26.99:30033 --token hdse4x.7d8vdil1s7tdp8dj --discovery-token-ca-cert-hash sha256:11bf0456c84c2a176eb380a98844eed80a296cdb6cc557c863991c63de594108 --control-plane --certificate-key 61b1a22aa5dbbd93660709aa343452d80ec98bf2efef6915b3a36ae97a8eefed --cri-socket /var/run/cri-dockerd.sock
......
┌──[root@vms102.liruilongs.github.io]-[~]
└─$kubeadm join 192.168.26.99:30033 --token hdse4x.7d8vdil1s7tdp8dj --discovery-token-ca-cert-hash sha256:11bf0456c84c2a176eb380a98844eed80a296cdb6cc557c863991c63de594108 --control-plane --certificate-key 61b1a22aa5dbbd93660709aa343452d80ec98bf2efef6915b3a36ae97a8eefed --cri-socket /var/run/cri-dockerd.sock
拷贝 kubeconfig 文件并 配置 命令补全。每个 master 节点执行。
mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config
source <(kubectl completion bash) >> /etc/profile
┌──[root@vms101.liruilongs.github.io]-[~]
└─$
工作节点添加
工作节点通过 ansible 并行批量添加。
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$ansible k8s_node -m shell -a "kubeadm join 192.168.26.99:30033 --token hdse4x.7d8vdil1s7tdp8dj --discovery-token-ca-cert-hash sha256:11bf0456c84c2a176eb380a98844eed80a296cdb6cc557c863991c63de594108 --cri-socket /var/run/cri-dockerd.sock " -i host.yaml
查看节点状态
┌──[root@vms100.liruilongs.github.io]-[~/ansible/calico]
└─$kubectl get nodes
NAME STATUS ROLES AGE VERSION
vms100.liruilongs.github.io Ready control-plane 17h v1.25.1
vms101.liruilongs.github.io Ready control-plane 15h v1.25.1
vms102.liruilongs.github.io Ready control-plane 15h v1.25.1
vms103.liruilongs.github.io Ready <none> 15h v1.25.1
vms105.liruilongs.github.io Ready <none> 15h v1.25.1
vms106.liruilongs.github.io Ready <none> 15h v1.25.1
vms107.liruilongs.github.io Ready <none> 15h v1.25.1
vms108.liruilongs.github.io Ready <none> 15h v1.25.1
┌──[root@vms100.liruilongs.github.io]-[~/ansible/calico]
└─$
到这里集群就算是安装完成，但是对于生产环境，我们需要安装一个常用的插件，集群备份，监控工具，Ingress 控制器等。这部分属于可选项。
部署后的可选操作
安装一些常用插件和工具
一些常用的工具和 kubelct 插件，这里我直接从旧的集群里拷贝过来。安装相对简单，没有网络的可以，下载二进制包，然后配置成 kubectl 插件。有的话可以先下载 krew ，通过 krew 下载其他的插件
┌──[root@vms100.liruilongs.github.io]-[/usr/local/bin]
└─$tree -h
.
├── [ 45M] helm
├── [ 15M] helmify
├── [9.2M] kubectl-ketall
├── [ 11M] kubectl-krew
├── [ 44M] kubectl-kubepug
├── [8.8M] kubectl-rakkess
├── [ 18M] kubectl-score
├── [ 57M] kubectl-spy
├── [ 31M] kubectl-tree
├── [ 57M] kubectl-virt
└── [ 14M] kustomize
0 directories, 11 files
工具介绍：
- helm: HELM chart 包管理器
- kustomize： YAML 文件整合管理工具，用于管理整合生成 资源 YAML文件
- helmify： YAML 文件转 HELM chart 包工具，可以把 YAML 文件转化为 charts 包
- kubectl-ketall： 查看所有集群资源的 kubelct 插件工具
- kubectl-krew： kubelet 插件管理工具，用于自动的安装升级 kubectl 插件。
- kubectl-kubepug： 集群API资源版本 查看，用于升级检查
- kubectl-rakkess： 集群 RBAC 权限查看工具，用于查看整个集群授权
- kubectl-score： API资源定义建议工具，用于给出一些 API 资源的优化建议。
- kubectl-spy： 集群 API 资源动态监控工具，可以看到具体的YAML 字段变动。
- kubectl-tree： 集群 API 资源 层级关系，用于展示 API 资源的 树状关系
- kubectl-virt: 集群虚机环境管理工具，用于管理接入集群中的虚拟机。
安装一些监控工具
cadvisor 安装
SPS 1.25不被支持，需要提前去掉，或者看看下官网的通过 kustomize 修改后安装
┌──[root@vms100.liruilongs.github.io]-[~/ansible/cadvisor]
└─$kubectl apply -f cadvisor.yaml
namespace/cadvisor created
serviceaccount/cadvisor created
clusterrole.rbac.authorization.k8s.io/cadvisor created
clusterrolebinding.rbac.authorization.k8s.io/cadvisor created
daemonset.apps/cadvisor created
安装 metrics-server 监控工具
┌──[root@vms100.liruilongs.github.io]-[~/ansible/cadvisor]
└─$kubectl apply -f metrics-server.yaml
serviceaccount/metrics-server created
clusterrole.rbac.authorization.k8s.io/system:aggregated-metrics-reader created
clusterrole.rbac.authorization.k8s.io/system:metrics-server created
rolebinding.rbac.authorization.k8s.io/metrics-server-auth-reader created
clusterrolebinding.rbac.authorization.k8s.io/metrics-server:system:auth-delegator created
clusterrolebinding.rbac.authorization.k8s.io/system:metrics-server created
service/metrics-server created
deployment.apps/metrics-server created
apiservice.apiregistration.k8s.io/v1beta1.metrics.k8s.io created
安装 ingress-nginx 用于 Ingress
为了让 Ingress 资源工作，集群必须有一个正在运行的 Ingress 控制器。
与作为 kube-controller-manager 可执行文件的一部分运行的其他类型的控制器不同， Ingress 控制器不是随集群自动启动的。 基于此页面，你可选择最适合你的集群的 ingress 控制器实现。
Kubernetes 作为一个项目，目前支持和维护 AWS、 GCE 和 Nginx Ingress 控制器
┌──[root@vms100.liruilongs.github.io]-[~/ansible/ingress_nginx]
└─$helm upgrade --install ingress-nginx ingress-nginx --repo https://kubernetes.github.io/ingress-nginx --namespace ingress-nginx --create-namespace
如果没有网，可以把 yaml 文件和 对应的 镜像导入进去。安装完成有个 Service 类型是 LB ,所有我们还的安装一个软 LB
┌──[root@vms100.liruilongs.github.io]-[~/ansible/ingress_nginx]
└─$kubectl apply -f deploy.yaml
namespace/ingress-nginx created
serviceaccount/ingress-nginx created
serviceaccount/ingress-nginx-admission created
role.rbac.authorization.k8s.io/ingress-nginx created
role.rbac.authorization.k8s.io/ingress-nginx-admission created
clusterrole.rbac.authorization.k8s.io/ingress-nginx created
clusterrole.rbac.authorization.k8s.io/ingress-nginx-admission created
rolebinding.rbac.authorization.k8s.io/ingress-nginx created
rolebinding.rbac.authorization.k8s.io/ingress-nginx-admission created
clusterrolebinding.rbac.authorization.k8s.io/ingress-nginx created
clusterrolebinding.rbac.authorization.k8s.io/ingress-nginx-admission created
configmap/ingress-nginx-controller created
service/ingress-nginx-controller created
service/ingress-nginx-controller-admission created
deployment.apps/ingress-nginx-controller created
job.batch/ingress-nginx-admission-create created
job.batch/ingress-nginx-admission-patch created
ingressclass.networking.k8s.io/nginx created
validatingwebhookconfiguration.admissionregistration.k8s.io/ingress-nginx-admission created
安装 Metallb 用于 LoadBalancer
Metallb 实现 LoadBalancer
Metallb可以通过k8s原生的方式提供LB类型的Service支持
kubectl apply -f https://raw.githubusercontent.com/metallb/metallb/v0.13.7/config/manifests/metallb-native.yaml
┌──[root@vms100.liruilongs.github.io]-[~/ansible/metallb]
└─$kubectl apply -f metallb-native.yaml
namespace/metallb-system created
customresourcedefinition.apiextensions.k8s.io/addresspools.metallb.io created
customresourcedefinition.apiextensions.k8s.io/bfdprofiles.metallb.io created
customresourcedefinition.apiextensions.k8s.io/bgpadvertisements.metallb.io created
customresourcedefinition.apiextensions.k8s.io/bgppeers.metallb.io created
customresourcedefinition.apiextensions.k8s.io/communities.metallb.io created
customresourcedefinition.apiextensions.k8s.io/ipaddresspools.metallb.io created
customresourcedefinition.apiextensions.k8s.io/l2advertisements.metallb.io created
serviceaccount/controller created
serviceaccount/speaker created
role.rbac.authorization.k8s.io/controller created
role.rbac.authorization.k8s.io/pod-lister created
clusterrole.rbac.authorization.k8s.io/metallb-system:controller created
clusterrole.rbac.authorization.k8s.io/metallb-system:speaker created
rolebinding.rbac.authorization.k8s.io/controller created
rolebinding.rbac.authorization.k8s.io/pod-lister created
clusterrolebinding.rbac.authorization.k8s.io/metallb-system:controller created
clusterrolebinding.rbac.authorization.k8s.io/metallb-system:speaker created
secret/webhook-server-cert created
service/webhook-service created
deployment.apps/controller created
daemonset.apps/speaker created
validatingwebhookconfiguration.admissionregistration.k8s.io/metallb-webhook-configuration created
创建 IP池
┌──[root@vms100.liruilongs.github.io]-[~/ansible/metallb]
└─$kubectl apply -f pool.yaml
ipaddresspool.metallb.io/first-pool created
┌──[root@vms100.liruilongs.github.io]-[~/ansible/metallb]
└─$cat pool.yaml
apiVersion: metallb.io/v1beta1
kind: IPAddressPool
metadata:
name: first-pool
namespace: metallb-system
spec:
addresses:
- 192.168.26.220-192.168.26.249
┌──[root@vms100.liruilongs.github.io]-[~/ansible/metallb]
└─$
┌──[root@vms100.liruilongs.github.io]-[~/ansible/metallb]
└─$cat l2a.yaml
apiVersion: metallb.io/v1beta1
kind: L2Advertisement
metadata:
name: example
namespace: metallb-system
┌──[root@vms100.liruilongs.github.io]-[~/ansible/metallb]
└─$kubectl apply -f l2a.yaml
l2advertisement.metallb.io/example unchanged
┌──[root@vms100.liruilongs.github.io]-[~/ansible/metallb]
└─$
安装一个本地存储的 SC
我们需要安装一个 SC 分配器，以后可能会用到，这个也可以以后安装 : https://github.com/rancher/local-path-provisioner
┌──[root@vms100.liruilongs.github.io]-[~/ansible/helm]
└─$kubectl apply -f local-path-storage.yaml
namespace/local-path-storage unchanged
serviceaccount/local-path-provisioner-service-account unchanged
clusterrole.rbac.authorization.k8s.io/local-path-provisioner-role unchanged
clusterrolebinding.rbac.authorization.k8s.io/local-path-provisioner-bind unchanged
deployment.apps/local-path-provisioner unchanged
storageclass.storage.k8s.io/local-path unchanged
configmap/local-path-config unchanged
安装 prometheus
需要指标监控，所有需要普罗米修斯
kube-prometheus-stack-30.0.1
https://github.com/prometheus-community/helm-charts/releases/download/kube-prometheus-stack-30.0.1/kube-prometheus-stack-30.0.1.tgz
镜像拉不了的问题，直接替换不好找，这里把 charts 包下载下来，然后通过 helm template 转化为具体的 yaml 文件。替换对应的镜像。但是这样还一个问题，一些 CRD 不会预先安装，尤其是多 master 的情况，这里你可以多试几次，说不定就可以了，github上有人提了，貌似没有很好的解决方案，我的解决办法是先用 helm 安装，然后卸载，卸载的时候不会卸载 crd，然后在运行 生成的 yaml 文件。
┌──[root@vms100.liruilongs.github.io]-[~/ansible/helm]
└─$cd kube-prometheus-stack
┌──[root@vms100.liruilongs.github.io]-[~/ansible/helm/kube-prometheus-stack]
└─$ls
Chart.lock charts Chart.yaml CONTRIBUTING.md crds README.md templates values.yaml
┌──[root@vms100.liruilongs.github.io]-[~/ansible/helm/kube-prometheus-stack]
└─$helm install kube-prometheus-stack .
┌──[root@vms100.liruilongs.github.io]-[~/ansible/helm/kube-prometheus-stack]
└─$helm template . > ../kube-prometheus-stack.yaml
执行应用
┌──[root@vms100.liruilongs.github.io]-[~/ansible/helm]
└─$kubectl apply -f kube-prometheus-stack.yaml
执行完需要修改svc 为 NodePort 当然如果有 Ingress 控制器，或者 LB ，可以配置其他类型。
┌──[root@vms100.liruilongs.github.io]-[~/ansible/helm]
└─$kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
alertmanager-operated ClusterIP None <none> 9093/TCP,9094/TCP,9094/UDP 27m
kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 40h
prometheus-operated ClusterIP None <none> 9090/TCP 27m
release-name-grafana NodePort 10.111.188.209 <none> 80:30203/TCP 30m
release-name-kube-promethe-alertmanager ClusterIP 10.97.17.175 <none> 9093/TCP 30m
release-name-kube-promethe-operator ClusterIP 10.107.60.174 <none> 443/TCP 30m
release-name-kube-promethe-prometheus ClusterIP 10.108.163.61 <none> 9090/TCP 30m
release-name-kube-state-metrics ClusterIP 10.102.37.208 <none> 8080/TCP 30m
release-name-prometheus-node-exporter ClusterIP 10.100.5.155 <none> 9100/TCP 30m
登录用户名和密码获取
┌──[root@vms100.liruilongs.github.io]-[~/ansible/helm]
└─$kubectl get secrets release-name-grafana -o jsonpath='{.data.admin-user}' | base64 -d
admin┌──[root@vms100.liruilongs.github.io]-[~/ansible/helm]
└─$kubectl get secrets release-name-grafana -o jsonpath='{.data.admin-password}' | base64 -d
prom-operator┌──[root@vms100.liruilongs.github.io]-[~/ansible/helm]
└─$
ETCD 快照文件定时备份
生产环境的 ETCD 一定要做备份，要不出了问题只能跑路了…
service 服务编写
这里我们写了一个脚本，通过 systemd.service 的方式运行。存放位置见注释
┌──[root@vms81.liruilongs.github.io]-[~/back]
└─$systemctl cat etcd-backup
# /usr/lib/systemd/system/etcd-backup.service
[Unit]
Description= "ETCD 备份"
After=network-online.target
[Service]
Type=oneshot
Environment=ETCDCTL_API=3
ExecStart=/usr/bin/bash /usr/lib/systemd/system/etcd_back.sh
[Install]
WantedBy=multi-user.target
定时任务编写
定时备份通过 systemd.timer 的方式实现
┌──[root@vms81.liruilongs.github.io]-[~/back]
└─$systemctl cat etcd-backup.timer
# /usr/lib/systemd/system/etcd-backup.timer
[Unit]
Description="每天备份一次 ETCD"
[Timer]
OnBootSec=3s
OnCalendar=*-*-* 00:00:00
Unit=etcd-backup.service
[Install]
WantedBy=multi-user.target
备份脚本编写
具体的 ETCD 快照备份脚本
┌──[root@vms81.liruilongs.github.io]-[~/back]
└─$cat /usr/lib/systemd/system/etcd_back.sh
#!/bin/bash
#@File : erct_break.sh
#@Time : 2023/01/27 23:00:27
#@Author : Li Ruilong
#@Version : 1.0
#@Desc : ETCD 备份
#@Contact : 1224965096@qq.com
if [ ! -d /root/back/ ];then
mkdir -p /root/back/
fi
STR_DATE=$(date +%Y%m%d%H%M)
ETCDCTL_API=3 etcdctl \
--endpoints="https://127.0.0.1:2379" \
--cert="/etc/kubernetes/pki/etcd/server.crt" \
--key="/etc/kubernetes/pki/etcd/server.key" \
--cacert="/etc/kubernetes/pki/etcd/ca.crt" \
snapshot save /root/back/snap-${STR_DATE}.db
ETCDCTL_API=3 etcdctl --write-out=table snapshot status /root/back/snap-${STR_DATE}.db
运行方式
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$systemctl enable etcd-backup.service --now
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$systemctl enable etcd-backup.timer --now
查看备份情况
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$ls -lh /root/back/
总用量 311M
-rw-r--r-- 1 root root 27M 1月 28 00:17 snap-202301280017.db
-rw-r--r-- 1 root root 27M 1月 29 00:00 snap-202301290000.db
-rw-r--r-- 1 root root 27M 2月 1 21:43 snap-202302012143.db
-rw-r--r-- 1 root root 27M 2月 2 00:00 snap-202302020000.db
-rw-r--r-- 1 root root 29M 2月 3 00:00 snap-202302030000.db
-rw-r--r-- 1 root root 29M 2月 4 00:00 snap-202302040000.db
-rw-r--r-- 1 root root 36M 2月 5 00:00 snap-202302050000.db
┌──[root@vms100.liruilongs.github.io]-[~/ansible]
└─$
安装 Velero 集群备份
Velero 用于集群 安全备份和恢复、执行灾难恢复以及迁移 Kubernetes 集群资源和持久卷。
客户端
┌──[root@vms100.liruilongs.github.io]-[~/ansible/velero]
└─$wget --no-check-certificate https://github.com/vmware-tanzu/velero/releases/download/v1.10.1-rc.1/velero-v1.10.1-rc.1-linux-amd64.tar.gz
┌──[root@vms100.liruilongs.github.io]-[~/ansible/velero]
└─$ls
velero-v1.10.1-rc.1-linux-amd64.tar.gz
┌──[root@vms100.liruilongs.github.io]-[~/ansible/velero]
└─$tar -zxvf velero-v1.10.1-rc.1-linux-amd64.tar.gz
┌──[root@vms100.liruilongs.github.io]-[~/ansible/velero]
└─$cd velero-v1.10.1-rc.1-linux-amd64/
┌──[root@vms100.liruilongs.github.io]-[~/ansible/velero/velero-v1.10.1-rc.1-linux-amd64]
└─$cp velero /usr/local/bin/
┌──[root@vms100.liruilongs.github.io]-[~/ansible/velero/velero-v1.10.1-rc.1-linux-amd64]
└─$velero version
Client:
Version: v1.10.1-rc.1
Git commit: e4d2a83917cd848e5f4e6ebc445fd3d262de10fa
<error getting server version: no matches for kind "ServerStatusRequest" in version "velero.io/v1">
服务端安装
配置登录相关的帐密文件
┌──[root@vms100.liruilongs.github.io]-[~/ansible/velero/velero-v1.10.1-rc.1-linux-amd64]
└─$vim credentials-velero
┌──[root@vms100.liruilongs.github.io]-[~/ansible/velero/velero-v1.10.1-rc.1-linux-amd64]
└─$cat credentials-velero
[default]
aws_access_key_id = minio
aws_secret_access_key = minio123
启动服务器和本地存储服务。在 Velero 目录中，在上面的客户端的安装包里，解压出来就可以看到
修改下 yaml 文件，这个YAM 文件在上面下载的 安装包里。
┌──[root@vms100.liruilongs.github.io]-[~/ansible/velero/velero-v1.10.1-rc.1-linux-amd64]
└─$cat examples/minio/00-minio-deployment.yaml
# Copyright 2017 the Velero contributors.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
---
apiVersion: v1
kind: Namespace
metadata:
name: velero
---
apiVersion: apps/v1
kind: Deployment
metadata:
namespace: velero
name: minio
labels:
component: minio
spec:
strategy:
type: Recreate
selector:
matchLabels:
component: minio
template:
metadata:
labels:
component: minio
spec:
volumes:
- name: storage
emptyDir: {}
- name: config
emptyDir: {}
containers:
- name: minio
image: quay.io/minio/minio:latest
imagePullPolicy: IfNotPresent
args:
- server
- /storage
- --console-address=:9090
- --config-dir=/config
env:
- name: MINIO_ROOT_USER
value: "minio"
- name: MINIO_ROOT_PASSWORD
value: "minio123"
ports:
- containerPort: 9000
- containerPort: 9090
volumeMounts:
- name: storage
mountPath: "/storage"
- name: config
mountPath: "/config"
---
apiVersion: v1
kind: Service
metadata:
namespace: velero
name: minio
labels:
component: minio
spec:
# ClusterIP is recommended for production environments.
# Change to NodePort if needed per documentation,
# but only if you run Minio in a test/trial environment, for example with Minikube.
type: NodePort
ports:
- port: 9000
name: api
targetPort: 9000
protocol: TCP
- port: 9099
name: console
targetPort: 9090
protocol: TCP
selector:
component: minio
---
apiVersion: batch/v1
kind: Job
metadata:
namespace: velero
name: minio-setup
labels:
component: minio
spec:
template:
metadata:
name: minio-setup
spec:
restartPolicy: OnFailure
volumes:
- name: config
emptyDir: {}
containers:
- name: mc
image: minio/mc:latest
imagePullPolicy: IfNotPresent
command:
- /bin/sh
- -c
- "mc --config-dir=/config config host add velero http://minio:9000 minio minio123 && mc --config-dir=/config mb -p velero/velero"
volumeMounts:
- name: config
mountPath: "/config"
不建议使用 empty dir 的方式。如果集群挂掉，很有可能无法重启 pod。以及对应的 容器。建议使用单独的容器部署挂载目录，
直接应用上面的 YAML 文件，访问查看。这里安装一个 minio 并且通过 job 添加了一个 桶，用于存放 备份后的数据。
通过命令行工具安装 velero
velero install \
--provider aws \
--plugins velero/velero-plugin-for-aws:v1.2.1 \
--bucket velero \
--secret-file ./credentials-velero \
--use-volume-snapshots=false \
--backup-location-config region=minio,s3ForcePathStyle="true",s3Url=http://minio.velero.svc:9000
- bucket：你在minio中创建的bucketname
- backup-location-config: 把xxx.xxx.xxx.xxx改成你minio服务器的ip地址。
也可以导出 YAML 文件在应用，确认没问题，或者私库需要替换相关镜像
┌──[root@vms100.liruilongs.github.io]-[~/ansible/velero/velero-v1.10.1-rc.1-linux-amd64]
└─$velero install \
--provider aws \
--plugins velero/velero-plugin-for-aws:v1.2.1 \
--bucket velero \
--secret-file ./credentials-velero \
--use-volume-snapshots=false \
--backup-location-config region=minio,s3ForcePathStyle="true",s3Url=http://minio.velero.svc:9000
--dry-run -o yaml > velero_deploy.yaml
┌──[root@vms100.liruilongs.github.io]-[~/ansible/velero/velero-v1.10.1-rc.1-linux-amd64]
└─$kubectl apply -f velero_deploy.yaml
CustomResourceDefinition/backuprepositories.velero.io: attempting to create resource
CustomResourceDefinition/backuprepositories.velero.io: attempting to create resource client
..........
BackupStorageLocation/default: attempting to create resource
BackupStorageLocation/default: attempting to create resource client
BackupStorageLocation/default: created
Deployment/velero: attempting to create resource
Deployment/velero: attempting to create resource client
Deployment/velero: created
Velero is installed! ⛵ Use 'kubectl logs deployment/velero -n velero' to view the status.
┌──[root@vms100.liruilongs.github.io]-[~/ansible/velero/velero-v1.10.1-rc.1-linux-amd64]
└─$
部署完成的 job 会自动新建
备份
┌──[root@vms100.liruilongs.github.io]-[~/ansible/velero/velero-v1.10.1-rc.1-linux-amd64]
└─$velero backup create velero-demo
Backup request "velero-demo" submitted successfully.
Run `velero backup describe velero-demo` or `velero backup logs velero-demo` for more details.
┌──[root@vms100.liruilongs.github.io]-[~/ansible/velero/velero-v1.10.1-rc.1-linux-amd64]
└─$velero get backup velero-demo
NAME STATUS ERRORS WARNINGS CREATED EXPIRES STORAGE LOCATION SELECTOR
velero-demo InProgress 0 0 2023-01-28 22:18:45 +0800 CST 29d default <none>
┌──[root@vms100.liruilongs.github.io]-[~/ansible/velero/velero-v1.10.1-rc.1-linux-amd64]
└─$
┌──[root@vms100.liruilongs.github.io]-[~/ansible/velero/velero-v1.10.1-rc.1-linux-amd64]
└─$velero get backup velero-demo
NAME STATUS ERRORS WARNINGS CREATED EXPIRES STORAGE LOCATION SELECTOR
velero-demo Completed 0 0 2023-01-28 22:18:45 +0800 CST 29d default <none>
篇幅原因，这块具体 velero 的 备份恢复,定时备份等不多说明。
关于 k8s 高可用集群部署就和小伙伴分享到这里，生活加油。
上面涉及到的 Ansible 剧本，静态pod yaml 文件，YUM配置文件，包括后来的一些常用工具安装的 yaml 文件都以整理上传 gitee， 小伙伴可自行下载，篇幅问题，没有展示。
获取git 仓库地址方式，关注 公总好 山河已无恙，回复 k8s-ha-deploy 即可获得地址。
博文部分内容参考
文中涉及参考链接内容版权归原作者所有，如有侵权请告知
https://kubernetes.io/zh-cn/docs/setup/production-environment/tools/
https://github.com/Mirantis/cri-dockerd
https://kubernetes.io/zh-cn/docs/setup/production-environment/container-runtimes/
https://github.com/kubernetes/kubeadm/blob/main/docs/ha-considerations.md#options-for-software-load-balancing
© 2018-2023 liruilonger@gmail.com, All rights reserved. 保持署名-非商用-相同方式共享(CC BY-NC-SA 4.0)
