Linux集群架构

Name Version
CentOS 7 3.10.0-693.el7.x86_64 #1 SMP Tue Aug 22 21:09:27 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux
Nginx nginx-1.12.1
keepalived keepalived.x86_64 0:1.3.5-6.el7

 

集群介绍

  • 根据功能划分为两大类:高可用和负载均衡 

  • 高可用集群通常为两台服务器,一台工作,另外一台作为冗余,当提供服务的机器宕机,冗余将接替继续提供服务

  • 实现高可用的开源软件有:heartbeat、keepalived

  • 负载均衡集群,需要有一台服务器作为分发器,它负责把用户的请求分发给后端的服务器处理,在这个集群里,除了分发器外,就是给用户提供服务的服务器了,这些服务器数量至少为2

  • 实现负载均衡的开源软件有LVS、keepalived、haproxy、nginx,商业的有F5、Netscaler

 

keepalived介绍

  • 在这里我们使用keepalived来实现高可用集群,因为heartbeat在centos6上有一些问题,影响实验效果

  • keepalived通过VRRP(Virtual Router Redundancy Protocl)来实现高可用。

  • 在这个协议里会将多台功能相同的路由器组成一个小组,这个小组里会有1个master角色和N(N>=1)个backup角色。

  • master会通过组播的形式向各个backup发送VRRP协议的数据包,当backup收不到master发来的VRRP数据包时,就会认为master宕机了。此时就需要根据各个backup的优先级来决定谁成为新的mater。

  • Keepalived要有三个模块,分别是core、check和vrrp。其中core模块为keepalived的核心,负责主进程的启动、维护以及全局配置文件的加载和解析,check模块负责健康检查,vrrp模块是来实现VRRP协议的。

 

用keepalived配置高可用集群

  • 准备两台机器128和130,128作为master,130作为backup

  • 两台机器都执行yum install -y keepalived

  • 两台机器都安装nginx,其中128上已经编译安装过nginx,130上需要yum安装nginx: yum install -y nginx

  • 设定vip为100,编辑128上keepalived配置文件,内容从https://coding.net/u/aminglinux/p/aminglinux-book/git/blob/master/D21Z/master_keepalived.conf获取

  [root@zyshanlinux-001 ~]# vim /etc/keepalived/keepalived.conf 
  ##快速清空文本的快捷方式##
  [root@zyshanlinux-001 ~]# > !$
  > /etc/keepalived/keepalived.conf
  [root@zyshanlinux-001 ~]# vim /etc/keepalived/keepalived.conf 
  global_defs {
     notification_email {
       [email protected]
     }
     notification_email_from [email protected]
     smtp_server 127.0.0.1
     smtp_connect_timeout 30
     router_id LVS_DEVEL
  }
  ​
  vrrp_script chk_nginx {
      script "/usr/local/sbin/check_ng.sh"
      interval 3
  }
  ​
  vrrp_instance VI_1 {
      state MASTER
      interface ens33
      virtual_router_id 51
      priority 100
      advert_int 1
      authentication {
          auth_type PASS
          auth_pass zyshanlinux>com
      }
      virtual_ipaddress {
          192.168.106.100
      }
      
      track_script {
          chk_nginx 
      }
      
  }       
  -- 插入 --

配置文件说明图

  • 130编辑监控脚本,内容从https://coding.net/u/aminglinux/p/aminglinux-book/git/blob/master/D21Z/master_check_ng.sh获取,给脚本755权限。

  ##检测脚本是/etc/keepalived/keepalived.conf里面定义的##
  [root@zyshanlinux-001 ~]# vim /usr/local/sbin/check_ng.sh
  #!/bin/bash
  #时间变量,用于记录日志
  d=`date --date today +%Y%m%d_%H:%M:%S`
  #计算nginx进程数量
  n=`ps -C nginx --no-heading|wc -l`
  #如果进程为0,则启动nginx,并且再次检测nginx进程数量,
  #如果还为0,说明nginx无法启动,此时需要关闭keepalived
  if [ $n -eq "0" ]; then
          /etc/init.d/nginx start
          n2=`ps -C nginx --no-heading|wc -l`
          if [ $n2 -eq "0"  ]; then
                  echo "$d nginx down,keepalived will stop" >> /var/log/check_ng.log
                  systemctl stop keepalived
          fi
  fi
  ​
  ​
  [root@zyshanlinux-001 ~]# chmod 755 /usr/local/sbin/check_ng.sh
  [root@zyshanlinux-001 ~]# ls -l !$
  ls -l /usr/local/sbin/check_ng.sh
  -rwxr-xr-x 1 root root 567 7月  22 21:52 /usr/local/sbin/check_ng.sh

  • 128启动keepalived服务

  [root@zyshanlinux-001 ~]# ps aux |grep keep
  root      1766  0.0  0.0 112720   972 pts/0    S+   21:57   0:00 grep --color=auto keep
  [root@zyshanlinux-001 ~]# systemctl start  keepalived
  [root@zyshanlinux-001 ~]# ps aux |grep keep
  root      1774  0.0  0.0 118652  1396 ?        Ss   21:57   0:00 /usr/sbin/keepalived -D
  root      1775  0.0  0.1 129580  3304 ?        S    21:57   0:00 /usr/sbin/keepalived -D
  root      1776  0.3  0.1 129520  2840 ?        R    21:57   0:00 /usr/sbin/keepalived -D
  root      1789  0.0  0.0 112720   972 pts/0    R+   21:57   0:00 grep --color=auto keep
  [root@zyshanlinux-001 ~]# ps aux |grep nginx
  root      1173  0.0  0.0  46032  1280 ?        Ss   21:10   0:00 nginx: master process /usr/local/nginx/sbin/nginx -c /usr/local/nginx/conf/nginx.conf
  nobody    1182  0.0  0.2  48520  4176 ?        S    21:10   0:00 nginx: worker process
  nobody    1183  0.0  0.2  48520  3920 ?        S    21:10   0:00 nginx: worker process
  root      1960  0.0  0.0 112720   968 pts/0    R+   21:58   0:00 grep --color=auto nginx

nginx服务即使停止也会自动启动。

  [root@zyshanlinux-001 ~]# /etc/init.d/nginx stop
  Stopping nginx (via systemctl):                            [  确定  ]
  [root@zyshanlinux-001 ~]# !ps
  ps aux |grep nginx
  root      2150  0.0  0.0  46032  1280 ?        Ss   22:00   0:00 nginx: master process /usr/local/nginx/sbin/nginx -c /usr/local/nginx/conf/nginx.conf
  nobody    2154  0.0  0.2  48520  3920 ?        S    22:00   0:00 nginx: worker process
  nobody    2155  0.0  0.2  48520  3920 ?        S    22:00   0:00 nginx: worker process
  root      2169  0.0  0.0 112720   972 pts/0    R+   22:00   0:00 grep --color=auto nginx

命令ip add才能看到100的IP,ifconfig就看不到了

检查防火墙是否关闭。

  [root@zyshanlinux-001 ~]# iptables -nvL
  Chain INPUT (policy ACCEPT 0 packets, 0 bytes)
   pkts bytes target     prot opt in     out     source               destination         
  ​
  Chain FORWARD (policy ACCEPT 0 packets, 0 bytes)
   pkts bytes target     prot opt in     out     source               destination         
  ​
  Chain OUTPUT (policy ACCEPT 0 packets, 0 bytes)
   pkts bytes target     prot opt in     out     source               destination         
  [root@zyshanlinux-001 ~]# getenforce
  Disabled

从上也要关闭防火墙。

  [root@zyshanlinux-02 ~]# systemctl stop firewalld
  [root@zyshanlinux-02 ~]# iptables -nvL
  Chain INPUT (policy ACCEPT 0 packets, 0 bytes)
   pkts bytes target     prot opt in     out     source               destination         
  ​
  Chain FORWARD (policy ACCEPT 0 packets, 0 bytes)
   pkts bytes target     prot opt in     out     source               destination         
  ​
  Chain OUTPUT (policy ACCEPT 0 packets, 0 bytes)
   pkts bytes target     prot opt in     out     source               destination         
  [root@zyshanlinux-02 ~]# getenforce
  Enforcing
  [root@zyshanlinux-02 ~]# setenforce 0
  [root@zyshanlinux-02 ~]# getenforce
  Permissive
  • 130上编辑配置文件,内容从https://coding.net/u/aminglinux/p/aminglinux-book/git/blob/master/D21Z/backup_keepalived.conf获取

  [root@zyshanlinux-02 ~]# vim /etc/keepalived/keepalived.conf 
  [root@zyshanlinux-02 ~]# > !$
  > /etc/keepalived/keepalived.conf
  [root@zyshanlinux-02 ~]# vim /etc/keepalived/keepalived.conf

  • 130上编辑监控脚本,内容从https://coding.net/u/aminglinux/p/aminglinux-book/git/blob/master/D21Z/backup_check_ng.sh获取 ,如果你是编译安装的脚本就要用/etc/init.d/nginx start,如果是yum安装的就要用systemctl start nginx,根据实际情况来。

  • 给脚本755权限,130上也启动服务 systemctl start keepalived

  [root@zyshanlinux-02 ~]# chmod 755 !$
  chmod 755 /usr/local/sbin/check_ng.sh
  [root@zyshanlinux-02 ~]# ls -l !$
  ls -l /usr/local/sbin/check_ng.sh
  -rwxr-xr-x. 1 root root 555 7月  22 22:30 /usr/local/sbin/check_ng.sh
  [root@zyshanlinux-02 ~]# systemctl start keepalived
  [root@zyshanlinux-02 ~]# ps aux |grep kee
  root      1590  0.0  0.0 118652  1400 ?        Ss   22:32   0:00 /usr/sbin/keepalived -D
  root      1591  0.0  0.1 127516  3304 ?        S    22:32   0:00 /usr/sbin/keepalived -D
  root      1592  0.1  0.1 127456  2844 ?        S    22:32   0:00 /usr/sbin/keepalived -D
  root      1631  0.0  0.0 112720   968 pts/0    R+   22:32   0:00 grep --color=auto kee
  [root@zyshanlinux-02 ~]# systemctl restart keepalived
  [root@zyshanlinux-02 ~]# ps aux |grep kee
  root      1711  0.0  0.0 118652  1392 ?        Ss   22:32   0:00 /usr/sbin/keepalived -D
  root      1712  0.0  0.1 127516  3292 ?        S    22:32   0:00 /usr/sbin/keepalived -D
  root      1713  0.3  0.1 127456  2836 ?        S    22:32   0:00 /usr/sbin/keepalived -D
  root      1723  0.0  0.0 112724   972 pts/0    S+   22:32   0:00 grep --color=auto kee

区分主从之间的nginx

主:源码包安装nginx的路径

  [root@zyshanlinux-001 ~]# cat /data/wwwroot/default/index.html
  master master.This is a default site.

从:yum安装nginx的路径

  [root@zyshanlinux-02 src]# cat /usr/share/nginx/html/index.html
  backup backup.

virtual_ipaddress:访问的是master,v-ip在master上

测试高可用

  • 先确定好两台机器上nginx差异,比如可以通过curl -I 来查看nginx版本

  • 测试1:关闭master上的nginx服务

无法关闭,keepalived会重新启动nginx。

  • 测试2:在master上增加iptabls规则 :iptables -I OUTPUT -p vrrp -j DROP

也无法测试出想要的效果,主服务继续启用没有宕机的效果。

  • 测试3:关闭master上的keepalived服务

主丢失v_ip,查看日志发现keepalived断开的问题

  [root@zyshanlinux-001 ~]# ip add
  1: lo:  mtu 65536 qdisc noqueue state UNKNOWN qlen 1
      link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
      inet 127.0.0.1/8 scope host lo
         valid_lft forever preferred_lft forever
      inet6 ::1/128 scope host 
         valid_lft forever preferred_lft forever
  2: ens33:  mtu 1500 qdisc pfifo_fast state UP qlen 1000
      link/ether 00:0c:29:a1:d4:eb brd ff:ff:ff:ff:ff:ff
      inet 192.168.106.128/24 brd 192.168.106.255 scope global ens33
         valid_lft forever preferred_lft forever
      inet 192.168.106.100/32 scope global ens33
         valid_lft forever preferred_lft forever
      inet 192.168.106.150/24 brd 192.168.106.255 scope global secondary ens33:0
         valid_lft forever preferred_lft forever
      inet6 fe80::8fc3:bbdf:ba89:22a7/64 scope link 
         valid_lft forever preferred_lft forever
  3: ens37:  mtu 1500 qdisc pfifo_fast state UP qlen 1000
      link/ether 00:0c:29:a1:d4:f5 brd ff:ff:ff:ff:ff:ff
  [root@zyshanlinux-001 ~]# systemctl stop keepalived
  [root@zyshanlinux-001 ~]# ip add
  1: lo:  mtu 65536 qdisc noqueue state UNKNOWN qlen 1
      link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
      inet 127.0.0.1/8 scope host lo
         valid_lft forever preferred_lft forever
      inet6 ::1/128 scope host 
         valid_lft forever preferred_lft forever
  2: ens33:  mtu 1500 qdisc pfifo_fast state UP qlen 1000
      link/ether 00:0c:29:a1:d4:eb brd ff:ff:ff:ff:ff:ff
      inet 192.168.106.128/24 brd 192.168.106.255 scope global ens33
         valid_lft forever preferred_lft forever
      inet 192.168.106.150/24 brd 192.168.106.255 scope global secondary ens33:0
         valid_lft forever preferred_lft forever
      inet6 fe80::8fc3:bbdf:ba89:22a7/64 scope link 
         valid_lft forever preferred_lft forever
  3: ens37:  mtu 1500 qdisc pfifo_fast state UP qlen 1000
      link/ether 00:0c:29:a1:d4:f5 brd ff:ff:ff:ff:ff:ff
  [root@zyshanlinux-001 ~]# less /var/log/messages
  Jul 22 23:54:26 zyshanlinux-001 Keepalived[1774]: Stopping
  Jul 22 23:54:26 zyshanlinux-001 systemd: Stopping LVS and VRRP High Availability Monitor...
  Jul 22 23:54:26 zyshanlinux-001 Keepalived_vrrp[1776]: VRRP_Instance(VI_1) sent 0 priority
  Jul 22 23:54:26 zyshanlinux-001 Keepalived_vrrp[1776]: VRRP_Instance(VI_1) removing protocol VIPs.
  Jul 22 23:54:26 zyshanlinux-001 Keepalived_healthcheckers[1775]: Stopped
  Jul 22 23:54:27 zyshanlinux-001 Keepalived_vrrp[1776]: Stopped
  Jul 22 23:54:27 zyshanlinux-001 systemd: Stopped LVS and VRRP High Availability Monitor.
  Jul 22 23:54:27 zyshanlinux-001 Keepalived[1774]: Stopped Keepalived v1.3.5 (03/19,2017), git commit v1.3.5-6-g6fa32f2

从监听v_ip,日志显示具体过程

  [root@zyshanlinux-02 src]# ip add
  1: lo:  mtu 65536 qdisc noqueue state UNKNOWN qlen 1
      link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
      inet 127.0.0.1/8 scope host lo
         valid_lft forever preferred_lft forever
      inet6 ::1/128 scope host 
         valid_lft forever preferred_lft forever
  2: ens33:  mtu 1500 qdisc pfifo_fast state UP qlen 1000
      link/ether 00:0c:29:29:ed:0e brd ff:ff:ff:ff:ff:ff
      inet 192.168.106.130/24 brd 192.168.106.255 scope global ens33
         valid_lft forever preferred_lft forever
      inet 192.168.106.100/32 scope global ens33
         valid_lft forever preferred_lft forever
      inet6 fe80::3c7e:461a:5056:da7d/64 scope link 
         valid_lft forever preferred_lft forever
  3: ens37:  mtu 1500 qdisc pfifo_fast state DOWN qlen 1000
      link/ether 00:0c:29:29:ed:18 brd ff:ff:ff:ff:ff:ff
      [root@zyshanlinux-02 src]# tail /var/log/messages
  Jul 22 23:54:31 zyshanlinux-02 Keepalived_vrrp[1733]: Sending gratuitous ARP on ens33 for 192.168.106.100
  Jul 22 23:54:31 zyshanlinux-02 Keepalived_vrrp[1733]: Sending gratuitous ARP on ens33 for 192.168.106.100
  Jul 22 23:54:31 zyshanlinux-02 Keepalived_vrrp[1733]: Sending gratuitous ARP on ens33 for 192.168.106.100
  Jul 22 23:54:31 zyshanlinux-02 Keepalived_vrrp[1733]: Sending gratuitous ARP on ens33 for 192.168.106.100
  Jul 22 23:54:36 zyshanlinux-02 Keepalived_vrrp[1733]: Sending gratuitous ARP on ens33 for 192.168.106.100
  Jul 22 23:54:36 zyshanlinux-02 Keepalived_vrrp[1733]: VRRP_Instance(VI_1) Sending/queueing gratuitous ARPs on ens33 for 192.168.106.100
  Jul 22 23:54:36 zyshanlinux-02 Keepalived_vrrp[1733]: Sending gratuitous ARP on ens33 for 192.168.106.100
  Jul 22 23:54:36 zyshanlinux-02 Keepalived_vrrp[1733]: Sending gratuitous ARP on ens33 for 192.168.106.100
  Jul 22 23:54:36 zyshanlinux-02 Keepalived_vrrp[1733]: Sending gratuitous ARP on ens33 for 192.168.106.100
  Jul 22 23:54:36 zyshanlinux-02 Keepalived_vrrp[1733]: Sending gratuitous ARP on ens33 for 192.168.106.100
  • 测试4:启动master上的keepalived服务

主:重新监听v_ip

  [root@zyshanlinux-001 ~]# systemctl start keepalived
  [root@zyshanlinux-001 ~]# ip add
  1: lo:  mtu 65536 qdisc noqueue state UNKNOWN qlen 1
      link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
      inet 127.0.0.1/8 scope host lo
         valid_lft forever preferred_lft forever
      inet6 ::1/128 scope host 
         valid_lft forever preferred_lft forever
  2: ens33:  mtu 1500 qdisc pfifo_fast state UP qlen 1000
      link/ether 00:0c:29:a1:d4:eb brd ff:ff:ff:ff:ff:ff
      inet 192.168.106.128/24 brd 192.168.106.255 scope global ens33
         valid_lft forever preferred_lft forever
      inet 192.168.106.100/32 scope global ens33
         valid_lft forever preferred_lft forever
      inet 192.168.106.150/24 brd 192.168.106.255 scope global secondary ens33:0
         valid_lft forever preferred_lft forever
      inet6 fe80::8fc3:bbdf:ba89:22a7/64 scope link 
         valid_lft forever preferred_lft forever
  3: ens37:  mtu 1500 qdisc pfifo_fast state UP qlen 1000
      link/ether 00:0c:29:a1:d4:f5 brd ff:ff:ff:ff:ff:ff
      inet6 fe80::7285:a690:d34:bb0c/64 scope link 
         valid_lft forever preferred_lft forever

从:丢失v_ip

  [root@zyshanlinux-02 src]# ip add
  1: lo:  mtu 65536 qdisc noqueue state UNKNOWN qlen 1
      link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
      inet 127.0.0.1/8 scope host lo
         valid_lft forever preferred_lft forever
      inet6 ::1/128 scope host 
         valid_lft forever preferred_lft forever
  2: ens33:  mtu 1500 qdisc pfifo_fast state UP qlen 1000
      link/ether 00:0c:29:29:ed:0e brd ff:ff:ff:ff:ff:ff
      inet 192.168.106.130/24 brd 192.168.106.255 scope global ens33
         valid_lft forever preferred_lft forever
      inet6 fe80::3c7e:461a:5056:da7d/64 scope link 
         valid_lft forever preferred_lft forever
  3: ens37:  mtu 1500 qdisc pfifo_fast state DOWN qlen 1000
      link/ether 00:0c:29:29:ed:18 brd ff:ff:ff:ff:ff:ff

 

负载均衡集群介绍

  • 主流开源软件LVS、keepalived、haproxy、nginx等

  • 其中LVS属于4层(网络OSI 7层模型),nginx属于7层,haproxy既可以认为是4层,也可以当做7层使用

  • keepalived的负载均衡功能其实就是lvs

  • lvs这种4层的负载均衡是可以分发除80外的其他端口通信的,比如MySQL的,而nginx仅仅支持http,https,mail,haproxy也支持MySQL这种

  • 相比较来说,LVS这种4层的更稳定,能承受更多的请求,而nginx这种7层的更加灵活,能实现更多的个性化需求

 

LVS介绍

  • LVS是由国人章文嵩开发

  • 流行度不亚于apache的httpd,基于TCP/IP做的路由和转发,稳定性和效率很高

  • LVS最新版本基于Linux内核2.6,有好多年不更新了

  • LVS有三种常见的模式:NAT、DR、IP Tunnel

  • LVS架构中有一个核心角色叫做分发器(Load balance),它用来分发用户的请求,还有诸多处理用户请求的服务器(Real Server,简称rs)

     

 

 

 

LVS的调度算法

  • 轮询 Round-Robin rr

  • 加权轮询 Weight Round-Robin wrr

  • 最小连接 Least-Connection lc

  • 加权最小连接 Weight Least-Connection wlc

  • 基于局部性的最小连接 Locality-Based Least Connections lblc

  • 带复制的基于局部性最小连接 Locality-Based Least Connections with Replication lblcr

  • 目标地址散列调度 Destination Hashing dh

  • 源地址散列调度 Source Hashing sh

 

LVS NAT模式搭建

三台机器 分发器,也叫调度器(简写为dir) 内网:133.130,外网:142.147(vmware仅主机模式) rs1 内网:133.132,设置网关为133.130 rs2 内网:133.133,设置网关为133.130 三台机器上都执行执行 systemctl stop firewalld; systemc disable firewalld systemctl start iptables-services; iptables -F; service iptables save

在dir上安装ipvsadm yum install -y ipvsdam 在dir上编写脚本,vim /usr/local/sbin/lvs_nat.sh//内容如下

  #! /bin/bash
  # director 服务器上开启路由转发功能
  echo 1 > /proc/sys/net/ipv4/ip_forward
  # 关闭icmp的重定向
  echo 0 > /proc/sys/net/ipv4/conf/all/send_redirects
  echo 0 > /proc/sys/net/ipv4/conf/default/send_redirects
  # 注意区分网卡名字,阿铭的两个网卡分别为ens33和ens37
  echo 0 > /proc/sys/net/ipv4/conf/ens33/send_redirects
  echo 0 > /proc/sys/net/ipv4/conf/ens37/send_redirects
  # director 设置nat防火墙
  iptables -t nat -F
  iptables -t nat -X
  iptables -t nat -A POSTROUTING -s 192.168.133.0/24  -j MASQUERADE
  # director设置ipvsadm
  IPVSADM='/usr/sbin/ipvsadm'
  $IPVSADM -C
  $IPVSADM -A -t 192.168.147.144:80 -s wlc -p 3
  $IPVSADM -a -t 192.168.147.144:80 -r 192.168.133.132:80 -m -w 1
  $IPVSADM -a -t 192.168.147.144:80 -r 192.168.133.133:80 -m -w 1
  ​

NAT模式效果测试

两台rs上都安装nginx 设置两台rs的主页,做一个区分,也就是说直接curl两台rs的ip时,得到不同的结果 浏览器里访问192.168.142.147,多访问几次看结果差异

 

 

 

拓展:

keepalived 默认报警设置使用外部邮箱http://blog.51cto.com/6764097/1954158

HAProxy+Keepalived配置邮件报警,会话保存机制和TCP端口范围(三)

https://blog.csdn.net/HzSunshine/article/details/62052398

LVS介绍及工作原理图解http://blog.51cto.com/jiekeyang/1839583

你可能感兴趣的:(linux)