生产环境k8s集群接入方案主要由以下三部分组成:
1、rancher ui,k8s可视化管理平台
2、nginx、k8s集群,业务服务访问入口
3、物理机
整体架构图:
准备:
1、所有主机节点开通外网访问权限
2、申请https://k8s-pro.xx.com rancher ui 域名解析
3、申请xxx.k8spro.xx.com 业务服务 域名解析
4、k8s相关命令文件地址:http://172.16.101.32/stgconfig/soft
一、安装环境
1、硬件环境
2、软件环境
os:centos7.6
rancher:2.3.2
docker:17.03.2-ce , overlay2存储,extfs 文件系统
二、安装前的准备
1、主机环境配置
1.1、 关闭防火墙
sudo systemctl stop firewalld.service && sudo systemctl disable firewalld.service
1.2、关闭setlinx
sudo sed -i 's/SELINUX=enforcing/SELINUX=disabled/g' /etc/selinux/config
1.3、关闭swap
注释或删除swap交换分区:vi /etc/fstab
#/dev/mapper/centos-swap swap swap defaults 0 0
1.4、配置主机时间、时区、系统语言
sudo ln -sf /usr/share/zoneinfo/Asia/Shanghai /etc/localtime
sudo echo 'LANG="en_US.UTF-8"' >> /etc/profile;source /etc/profile
1.5、修改最大进程数 、 用户进程打开文件数
echo '1000000' > /proc/sys/kernel/pid_max
vim /etc/sysctl.conf
kernel.pid_max=1000000
sysctl -p 保存生效
--------
vi /etc/security/limits.conf
* soft nproc 102400
* hard nofile 65535
jyapp hard nofile 102400
jyapp soft nofile 102400
jyapp soft nproc 102400
root soft nproc unlimited
root soft nofile 102400
root hard nofile 102400
1.6、Kernel性能调优
vi /etc/sysctl.conf
追加
net.ipv4.ip_forward=1
net.bridge.bridge-nf-call-iptables=1
net.ipv4.neigh.default.gc_thresh1=4096
net.ipv4.neigh.default.gc_thresh2=6144
net.ipv4.neigh.default.gc_thresh3=8192
1.7、修改本机dns
vi /etc/resolv.conf
nameserver 172.16.101.28
nameserver 172.16.101.18
2、安装docker-ce并配置
# 添加用户(可选)
sudo adduser jyapp
# 为新用户设置密码
sudo passwd jyapp
# 为新用户添加sudo权限
sudo echo 'jyapp ALL=(ALL) ALL' >> /etc/sudoers
Docker 9步 快速安装(https://blog.csdn.net/chen978616649/article/details/80505195)
jyapp加用docker用户组
sudo usermod -a -G docker jyapp
vi /etc/docker/daemon.json
{
"bip":"172.25.100.1/24"
,"insecure-registries":["172.16.101.43:5000","172.16.101.49","10.50.182.85","docker.xx.com"]
,"registry-mirrors": ["http://172.16.101.49","http://10.50.182.85","http://docker.xx.com","https://088aunhx.mirror.aliyuncs.com"]
,"log-driver": "json-file"
,"log-opts": {
"max-size": "100m",
"max-file": "3"
}
,"max-concurrent-downloads": 10
,"max-concurrent-uploads": 10
,"storage-driver": "overlay2"
,"storage-opts": ["overlay2.override_kernel_check=true"]
}
systemctl daemon-reload
systemctl restart docker
3、下载命令行工具
附件中下载
kubectl,rke,helm 到 172.19.101.135 /usr/local/bin 下
命令工具下载地址:http://172.16.101.32/stgconfig/soft/raw/master/k8s-kub.tar.gz
cd /usr/local/bin
赋可执行权限
sudo chmod +x *
修改profile
sudo vi /etc/profile
追加:
export PATH="/usr/local/bin:$PATH"
保存退出,使之生效
sudo source /etc/profile
4、免密登录
在172.19.101.135用jyapp用户执行:
ssh-keygen -t rsa
ssh-copy-id -p 10022 -i ~/.ssh/id_rsa.pub [email protected]
ssh-copy-id -p 10022 -i ~/.ssh/id_rsa.pub [email protected]
ssh-copy-id -p 10022 -i ~/.ssh/id_rsa.pub [email protected]
三、rke安装k8s
1、配置rke 创建rancher-cluster.yml文件
nodes:
- address: 172.19.101.135
user: jyapp
role: [controlplane,worker,etcd]
port: 10022
- address: 172.19.101.136
user: jyapp
role: [controlplane,worker,etcd]
port: 10022
- address: 172.19.101.137
user: jyapp
role: [controlplane,worker,etcd]
port: 10022
services:
etcd:
backup_config:
enabled: true
interval_hours: 6
retention: 12
extra_args:
quota-backend-bytes: '4294967296'
kubelet:
cluster_domain: cluster.local
cluster_dns_server: 10.43.0.10
fail_swap_on: true
extra_args:
max-pods: 50
kube-api-burst: '100'
kube-api-qps: '100'
#enforce-node-allocatable: 'pods'
#system-reserved: 'cpu=1,memory=4000Mi'
#kube-reserved: 'cpu=1,memory=4000Mi'
kube-controller:
cluster_cidr: 10.42.0.0/16
service_cluster_ip_range: 10.43.0.0/16
extra_args:
node-cidr-mask-size: '24'
kube-api:
service_cluster_ip_range: 10.43.0.0/16
service_node_port_range: 30000-32767
pod_security_policy: false
always_pull_images: false
2、运行RKE命令
rke up --config rancher-cluster.yml
完成后,它应显示:Finished building Kubernetes cluster successfully。
在本目录会生成config文件:kube_config_rancher-cluster.yml
如若操作失败,重新安装需要清理数据:
su - root
rm -rf /var/lib/rancher/etcd/*
rm -rf /etc/kubernetes/*
su - jyapp
rke remove --config ./rancher-cluster.yml
3、测试集群
mkdir ~/.kube
cp kube_config_rancher-cluster.yml ~/.kube/config
kubectl get cs
NAME STATUS MESSAGE ERROR
controller-manager Healthy ok
scheduler Healthy ok
etcd-0 Healthy {"health": "true"}
etcd-2 Healthy {"health": "true"}
etcd-1 Healthy {"health": "true"}
kubectl get node
NAME STATUS ROLES AGE VERSION
172.19.101.135 Ready controlplane,etcd,worker 20d v1.13.5
172.19.101.136 Ready controlplane,etcd,worker 20d v1.13.5
172.19.101.137 Ready controlplane,etcd,worker 20d v1.13.5
四、安装 rancher server
1、分别在172.19.101.117、172.19.101.134上执行
docker run -d --restart=unless-stopped \
-p 80:80 -p 443:443 \
-v /var/lib/rancher/:/var/lib/rancher/ \
-v /root/var/log/auditlog:/var/log/auditlog \
-e CATTLE_SYSTEM_CATALOG=bundled \
-e AUDIT_LEVEL=3 \
rancher/rancher:v2.3.2
访问:https://172.19.101.117
注:ui初始登录时需指访问地址为 k8s-pro.xx.com
2、通过rancher ui导入 rke k8s集群
五、 配置nginx
1、sudo vi /etc/nginx/nginx.conf
user nginx;
worker_processes 4;
error_log /var/log/nginx/error.log warn;
worker_rlimit_nofile 40000;
events {
worker_connections 8192;
}
http {
upstream rancher_web{
server 172.19.101.135:80;
server 172.19.101.136:80;
server 172.19.101.137:80;
}
server {
listen 80 default_server;
server_name _;
access_log logs/busiweb.access.log main;
client_max_body_size 300m;
client_body_buffer_size 128k;
proxy_connect_timeout 600;
proxy_read_timeout 600;
proxy_send_timeout 600;
proxy_buffer_size 64k;
proxy_buffers 4 32k;
proxy_busy_buffers_size 64k;
proxy_temp_file_write_size 64k;
location / {
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_pass http://rancher_web;
}
}
}
2、 或可以容器化nginx代替以上安装
下载附件nginx.tar.gz 并解压
略作修改:按上面nginx.conf 修改
执行 sh startNginx-80.sh 即可
六、jdk主机安装
集群各节点 135、136、137均需安装
jdk下载地址:
jdk为开发、测试、准生产、生产环境通用的标准及版本1.7和1.8,支持中文字体
下载方式:wget http://172.16.101.32/stgconfig/soft/raw/master/jdk.tar.gz
解压方式:tar -zxf jdk.tar.gz
解压到 /root/目录下即可
七、日志收集 elk
八、监控 promethus
九、jenkins 发版
十、需开通网络访问权限
1、镜像仓库 172.16.101.49,10.50.182.85
2、apollo
十一、新旧系统对接
如上架构图
直接在原有服务域名负载中追加
172.19.101.115:80;
172.19.101.116:80;
即可。