K8s(v1.25.1) 高可用集群(3 Master + 5 Node) Ansible 剧本部署
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: -
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: -
kubectl
常用插件安装 -
cadvisor
监控工具部署 -
metrics-server
监控工具部署 -
Ingress—nginx
Ingress 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]
└─$kubectl get nodes
NAME STATUS ROLES AGE VERSION
vms100.liruilongs.github.io Ready control-plane 10d v1.25.1
vms101.liruilongs.github.io Ready control-plane 10d v1.25.1
vms102.liruilongs.github.io Ready control-plane 10d v1.25.1
vms103.liruilongs.github.io Ready <none> 10d v1.25.1
vms105.liruilongs.github.io Ready <none> 10d v1.25.1
vms106.liruilongs.github.io Ready <none> 10d v1.25.1
vms107.liruilongs.github.io Ready <none> 10d v1.25.1
vms108.liruilongs.github.io Ready <none> 10d v1.25.1
部署涉及机器
-
master1: 192.168.26.100 -
master2: 192.168.26.101 -
master3: 192.168.26.102 -
work Node1: 192.168.26.103 -
work Node2: 192.168.26.105 -
work Node3: 192.168.26.106 -
work Node4: 192.168.26.107 -
work Node5: 192.168.26.108
哈 104 不吉利,跳过去了.....
集群部署
部署拓扑
下图为 k8s 官网文档中 HA 拓扑图,这里使用下面的拓扑方式部署
K8s 官网HA 拓扑图堆叠(Stacked)HA 集群拓扑, 其中 etcd 分布式数据存储集群堆叠在 kubeadm 管理的控制平面节点上
,作为控制平面的一个组件运行。
部署后的 etcd 分布
┌──[root@vms100.liruilongs.github.io]-[~/ansible/kubescape]
└─$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
。便于LEADER
选取
┌──[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 gitee 仓库,获取 git 仓库地址方式见文末
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仓库 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,对于生产环境,我们可能需要安装一些常用的插件,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]
└─$tree -h
.
├── [ 45M] helm
├── [ 15M] helmify
├── [9.2M] kubectl-ketall
├── [ 11M] kubectl-krew
├── [ 44M] kubectl-kubepug
├── [8.8M] kubectl-rakkess
├── [ 18M] kubectl-score
├── [ 57M] kubectl-spy
├── [ 31M] kubectl-tree
├── [ 57M] kubectl-virt
└── [ 14M] kustomize
0 directories, 11 files
工具介绍:
-
helm: HELM chart 包管理器 -
kustomize: YAML 文件整合管理工具,用于管理整合生成 资源 YAML文件 -
helmify: YAML 文件转 HELM chart 包工具,可以把 YAML 文件转化为 charts 包 -
kubectl-ketall: 查看所有集群资源的 kubelct 插件工具 -
kubectl-krew: kubelet 插件管理工具,用于自动的安装升级 kubectl 插件。 -
kubectl-kubepug: 集群API资源版本 查看,用于升级检查 -
kubectl-rakkess: 集群 RBAC 权限查看工具,用于查看整个集群授权 -
kubectl-score: API资源定义建议合规检查工具,用于给出一些 API 资源的优化建议。 -
kubectl-spy: 集群 API 资源动态监控工具,可以看到具体的YAML 字段变动。 -
kubectl-tree: 集群 API 资源 层级关系,用于展示 API 资源的 树状关系 -
kubectl-virt: 集群虚机环境管理工具,用于管理接入集群中的虚拟机。
安装一些监控工具
cadvisor 安装
SPS 1.25不被支持,需要提前去掉,或者看看下官网的通过 kustomize
修改后安装
┌──[root@vms100.liruilongs.github.io]-[~/ansible/cadvisor]
└─$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
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]
└─$kubectl apply -f local-path-storage.yaml
namespace/local-path-storage unchanged
serviceaccount/local-path-provisioner-service-account unchanged
clusterrole.rbac.authorization.k8s.io/local-path-provisioner-role unchanged
clusterrolebinding.rbac.authorization.k8s.io/local-path-provisioner-bind unchanged
deployment.apps/local-path-provisioner unchanged
storageclass.storage.k8s.io/local-path unchanged
configmap/local-path-config unchanged
安装 prometheus
需要指标监控,所有需要普罗米修斯
helm 方式安装 kube-prometheus-stack-30.0.1
https://github.com/prometheus-community/helm-charts/releases/download/kube-prometheus-stack-30.0.1/kube-prometheus-stack-30.0.1.tgz
镜像拉不了的问题,直接替换不好找,这里把 charts 包下载下来,然后通过 helm template
转化为具体的 yaml 文件。替换对应的镜像。但是这样还一个问题,一些 CRD 不会预先安装,尤其是多 master 的情况,这里你可以多试几次,说不定就可以了,github上有人提了,貌似没有很好的解决方案,我的解决办法是先用 helm 安装,然后卸载,卸载的时候不会卸载 crd,然后在运行 生成的 yaml 文件。
┌──[root@vms100.liruilongs.github.io]-[~/ansible/helm]
└─$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>
在这里插入图片描述
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-自由转载-相同方式共享(创意共享 3.0 许可证)
- 点赞
- 收藏
- 关注作者
评论(0)