keepalived配置

摘要： 給出的虛擬IP為10.0.0.110 給出MASTER的配置檔案： ! Configuration File for keepalived #這裡給出keepalived -v的版本資訊 #Keepalived v1.3.5 (0...

給出的虛擬IP為10.0.0.110

1. 給出MASTER的配置檔案：

   ! Configuration File for keepalived
   #這裡給出keepalived -v的版本資訊
   #Keepalived v1.3.5 (03/19,2017), git commit v1.3.5-6-g6fa32f2
   global_defs {
   router_id nginx_load_balance_01
   enable_script_security on
   script_user keepalived_script
   #不對，原因未知
   #default_interface ens33
   }
   #注意，這裡一定要給成雙引號，單引號不行
   vrrp_script check_nginx {
   script "/home/keepalived_script/check_nginx"
   weight -7
   }
   
   vrrp_instance nginx_load_balance {
   state MASTER
   interface ens33
   virtual_router_id 51
   priority 10
   advert_int 1
   #nopreempt好像得配置在backup機器上
   #nopreempt 
   authentication {
   auth_type PASS
   auth_pass 111111
   }
   #不好使__start__
   #指定要track哪些網絡卡，一旦這些網絡卡down了，那麼，這個vrrp_instance就進入fault狀態
   #track_interface {
   #ens33
   #ens34
   #}
   #不好使__end__
   #因為只是在網路層面track網絡卡是不夠的，我們還要進行監控某些應用層的東西
   track_script {
   check_nginx
   }
   #注意，這裡一定要給成雙引號，單引號不行
   notify_master "/home/keepalived_script/change_to_master"
   notify_backup "/home/keepalived_script/change_to_backup"
   notify_fault"/home/keepalived_script/change_to_fault"
   virtual_ipaddress {
   10.0.0.110/8
   }
   
   }

2. 然後是/home/keepalived_script/check_nginx 指令碼：

   #!/bin/env sh
   ret=$(ss -tunlp | grep 80 | wc -l)
   if [ $ret -lt 1 ]; then
   exit 1
   else 
   exit 0
   fi

3. 然後是BACKUP的配置：

   ! Configuration File for keepalived
   #這裡給出keepalived -v的版本資訊 
   #Keepalived v1.3.5 (03/19,2017), git commit v1.3.5-6-g6fa32f2
   global_defs {
   router_id nginx_load_balance_02
   enable_script_security on
   script_user keepalived_script
   }
   #注意，這裡一定要使用雙引號
   vrrp_script check_nginx {
   script "/home/keepalived_script/check_nginx"
   weight 7
   }
   vrrp_instance nginx_load_balance {
   state BACKUP
   interface ens33
   virtual_router_id 51
   priority 5
   advert_int 1
   #這個不能開啟。這個選項的意思是當MASTER down了以後，
   #BACKUP不搶佔虛擬IP。
   #啟動時，日誌輸出中有Preempt disabled
   #nopreempt on
   authentication {
   auth_type PASS
   auth_pass 111111
   }
   #不好使__start__
   #指定要track哪些網絡卡，一旦這些網絡卡down了，那麼，這個vrrp_instance就進入fault狀態
   #track_interface {
   #ens33
   #ens34
   #}
   #不好使__end__
   #注意，這裡一定要給成雙引號，單引號不行
   notify_master "/home/keepalived_script/change_to_master"
   notify_backup "/home/keepalived_script/change_to_backup"
   notify_fault"/home/keepalived_script/change_to_fault"
   #因為只是在網路層面track網絡卡是不夠的，我們還要進行監控某些應用層的東西
   track_script {
   check_nginx
   }
   virtual_ipaddress {
   10.0.0.110/8
   }
   }

4. 最後是/home/keepalived_script/check_nginx 指令碼：

   #!/bin/sh
   ret=$(ss -tunlp | grep 80 | wc -l)
   if [ $ret -lt 1 ]; then
   exit 0
   else 
   exit 1
   fi

5. 遺留問題：

   1. global_defs中的default_interface什麼意思，會在哪個模組中用到，VRRP，LVS還是SNMP？
   2. 啟動時，日誌檔案（/var/log/messages）中會有這麼一個報錯：“PID file /var/run/keepalived.pid not readable (yet?) after start.”，這是為什麼？這裡給出/usr/lib/systemd/system/keepalived.service中的配置資訊：

      [Unit]
      Description=LVS and VRRP High Availability Monitor
      After=syslog.target network-online.target
      
      [Service]
      Type=forking
      PIDFile=/var/run/keepalived.pid
      KillMode=process
      EnvironmentFile=-/etc/sysconfig/keepalived
      ExecStart=/usr/sbin/keepalived $KEEPALIVED_OPTIONS
      ExecReload=/bin/kill -HUP $MAINPID
      
      [Install]
      WantedBy=multi-user.target