corosync / pacemaker 기반의 DB이중화 구현

1. DB이중화 구성
   
   
   
2. DB 모니터링을  위한 스크립트 설치
   
   1

   $> vi /etc/init.d/db_check

   2

   #!/bin/bash

   3

   4

   case "$1" in

   5

    "start")

   6

      echo "start"

   7

    ;;

   8

    "stop")

   9

      echo "stop"

   10

    ;;

   11

    "status")

   12

     mysql --connect-timeout 2 -u monitor -p'모니터 계정패스워드' -h localhost -Nse 'select now()'

   13

     result_code=$?

   14

   15

     if [[ $result_code -eq 1 ]]

   16

     then

   17

      echo "[$(date +%Y-%m-%d-%H:%M:%S)] DB Connection Error" >> /var/log/db_check

   18

      pcs resource move DB_GROUP {{ 상대방IP }}

   19

     fi

   20

    ;;

   21

    *)

   22

      echo "start|stop|status"

   23

      exit 1

   24

    ;;

   25

   esac
   
   

3. DB이중화 구성을 위한 리소스 설정
   

   1

   $> pcs cluster setup db_ha {{ DB1_IP }} {{ DB2_IP }} --force

   2

   $> pcs cluster start --all

   3

   $> pcs property set stonith-enabled=false

   4

   $> pcs property set no-quorum-policy=ignore

   5

   $> pcs resource defaults update resource-stickiness=100

   6

   $> pcs resource create DB_CHECK service:db_check op status timeout=10s interval=10s --group DB_GROUP

   7

   $> pcs resource create DB_VIP  ocf:heartbeat:IPaddr2 ip={{ DB_VIP }} cidr_netmask=32 op monitor interval=10s --group DB_GROUP