1 备份CRS环境 root用户下执行
cd /u01
tar –cvf app.tar app
2 更改CRS配置
cd /u01/app/11.2.0/grid/bin
grid@prdweb1:/u01/app/11.2.0/grid/bin$oifcfg iflist
eth1 168.160.60.0
eth2 10.11.11.0
eth2 169.254.0.0
eth3 10.11.12.0
eth3 169.254.64.0
eth6 10.11.13.0
eth6 169.254.128.0
eth7 10.11.14.0
eth7 169.254.192.0
bond0 192.168.15.0
./oifcfg delif –globaleth1/168.160.60.0
./oifcfg delif –globalbond0/192.168.15.0
./oifcfg setif –global bond0/168.160.60.0
3 更改操作系统设置
3.1 更改/etc/hosts中节点1的IP地址列表
#bon0 Public Network
#eth0-4 Public Network
168.160.60.88 prdweb1
168.160.60.91 prdweb2
#Public Virtual IP (VIP) addresses
168.160.60.89 prdweb1-vip
168.160.60.92 prdweb2-vip
#SCAN For Oracle Cluster
168.160.60.90 prdweb-cluster-scan
3.2 完成节点1到节点2的互信验证,从节点2ssh hosts中的域名prdweb1
3.3 更改/etc/sysconfig/network-scripts/中的文件,更改IP地址。
3.4 联系网络组进行跳线,之后 servicenetwork restart
4 更改VIP资源
4.1 查看当前VIP配置
srvctl config nodeapps -a
grid@prdweb1:/u01/app/11.2.0/grid/bin$srvctl confignodeapps -a
Networkexists: 1/192.168.15.0/255.255.255.0/bond0, type static
VIPexists: /prdweb1-vip/192.168.15.89/192.168.15.0/255.255.255.0/bond0, hostingnode prdweb1
VIPexists: /prdweb2-vip/192.168.15.92/192.168.15.0/255.255.255.0/bond0, hostingnode prdweb2
crsctl stat res –t 应该显示VIP在线
4.2 停掉VIP资源 grid 用户执行
$ srvctl stop instance -d prdweb -i prdweb1
$ srvctl stop vip -n prdweb1 -f
crs_stat –t
4.3 更改VIP root用户执行
srvctl modify nodeapps -n prdweb1 -A prdweb1-vip/255.255.255.0/bond0
srvctl config nodeapps –a
4.4 重启资源 grid用户执行
$ srvctl start vip -n prdweb1
$ srvctl start listener -n prdweb1
$ srvctl start instance -d prdweb -n prdweb1
crs_stat –t –v
5 重复以上步骤更改2节点
6 更改scan-ip
6.1 查看状态,root用户执行
srvctl config scan
6.2 停止监听和scan,root用户执行
srvctl stop scan_listener
srvctl stop scan
6.3 重新配置scan-ip root用户执行
srvctl modify scan –n prdweb-cluster-scan
srvctl config scan
6.4 重启scan和scan监听 root用户执行
srvctl start scan
srvctl start scan_listener
6.5 如果scan IP的数目发生辩护,执行以下更新:
srvctl modify scan_listener -u
详细参见文档276434.1