oracle主、从自动切换方案

各个部分功能说明：

1、keepalived只用作vip功能；

2、check_oracle.sh脚本用于检测master服务，并执行vip、oracle切换；

通过“select status from v$instance;”获取master上oracle的数据状态，由返回值进行判断。

check_oracle.sh

#!/bin/sh

monitor_ip=192.168.3.186

ssh_port=22

logfile=/etc/keepalived/check_oracle.log

#check ssh port status

function check_ssh_port()

{

/usr/bin/nc -z -w 5 -n $monitor_ip $ssh_port 1 > /dev/nul

if [ $? = 0 ]; then

ssh_status=0

else

ssh_status=1

fi

return $ssh_status

}

#oracle_check_status

function check_oracle_helth()

{

source /home/oracle/.bash_profile

source /home/oracle/.bashrc

oracle_check_status=$(su - oracle -c "sqlplus sys/oracle@${monitor_ip}:1521/mydb as sysdba | grep -A 2 STATUS | sed -n '3p'" <

select status from v$instance;

EOF

)

if [ "$oracle_check_status" == OPEN ];then

oracle_status=0

else

oracle_status=1

fi

#echo $oracle_check_status

return $oracle_status

}

#update oracle slave to master

function change_oracle_status()

{

source /home/oracle/.bash_profile

source /home/oracle/.bashrc

su - oracle -c 'sqlplus / as sysdba' <

alter database recover managed standby database finish force;

alter database commit to switchover to primary with session shutdown;

alter database open;

EOF

}

check_time=5

#!=0

while [ $check_time -ne 0 ];do

let "check_time -= 1"

check_ssh_port

echo $ssh_status

if [ $ssh_status = 0 ];then

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - SSH ok, success_code:[$ssh_status]

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - start check oracle status......

check_oracle_helth

if [ $oracle_status = 0 ];then

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - ORACLE ok, success_code:[$oracle_status]

else

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - ORACLE Down, fail_code:[$oracle_status]

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - Keepalived stop......

ssh root@$monitor_ip '/etc/init.d/keepalived stop'

wait

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - Keepalived Already stop

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - Start update oracle slave stauts......

change_oracle_status

wait

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - Oracle slave update to master Success

sed 's/.*check_oracle.sh$//g' /var/spool/cron/root

/etc/init.d/crond restart

fi

else

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - SSH error 1, fail_code:[$ssh_status]

sleep 3

check_ssh_port

if [ $ssh_status = 0 ];then

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - SSH ok, success_code:[$ssh_status]

else

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - SSH error 2, fail_code:[$ssh_status]

sleep 3

check_ssh_port

if [ $ssh_status = 0 ];then

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - SSH ok, success_code:[$ssh_status]

else

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - SSH error 3, fail_code:[$ssh_status]

sleep 3

check_ssh_port

if [ $ssh_status = 0 ];then

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - SSH ok, success_code:[$ssh_status]

else

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - SSH error 4, fail_code:[$ssh_status]

sleep 3

check_ssh_port

if [ $ssh_status = 0 ];then

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - SSH ok, success_code:[$ssh_status]

else

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - SSH error 5, fail_code:[$ssh_status]

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - Start update oracle slave stauts......

change_oracle_status

wait

echo `date --date=today +"%Y-%M-%d %H:%m:%S"` - [INFO] - Oracle slave update to master Success

sed 's/.*check_oracle.sh$//g' /var/spool/cron/root

/etc/init.d/crond restart

fi

fi

fi

fi

fi

sleep 5

done

检测流程图：

脚本由crond执行，每1分钟执行一次，每执行一次，脚本内循环检测5次，首先判断服务器ssh端口是否正常，正常情况下，再去检查oracle状态，根据返回的状态决定是否进行切换，如果ssh端口异常，则进行5次检查判断，5次以内返回正常的话，则重新进行检测，如果第5次仍异常，则进行切换。