ORACLE RAC11G changes IP address
Environment: oracle 11.2.0.4 rac
IP address before modification
# public ip
192.168.2.71 db1
192.168.2.72 db2
192.168.2.76 db3
# priv ip
200.100.100.11 db1-priv
200.100.100.12 db2-priv
200.100.100.13 db3-priv
# vip ip
192.168.2.73 db1-vip
192.168.2.74 db2-vip
192.168.2.77 db3-vip
# scan ip
192.168.2.75 db-scan
Modified IP address
# public ip
192.168.1.71 db1
192.168.1.72 db2
192.168.1.76 db3
# priv ip
100.100.100.11 db1-priv
100.100.100.12 db2-priv
100.100.100.13 db3-priv
# vip ip
192.168.1.73 db1-vip
192.168.1.74 db2-vip
192.168.1.77 db3-vip
# scan ip
192.168.1.75 db-scan
1. View the backup of ocr
[root@db1 bin] #. / ocrconfig-showbackup
2. Backup OCR
[root@db1 bin] #. / ocrconfig-manualbackup
Db3 2015-01-29 19:58:22 / u01/app/11.2.0/grid/cdata/db-scan/backup_20150129_195822.ocr
3. Close the database
Srvctl stop database-d db
4. Modify on any node:
Here I operate on Node 1.
[root@db1 ~] # cd / u01/app/11.2.0/grid/bin/
[root@db1 bin] #. / oifcfg getif
Eth0 192.168.2.0 global public
Eth2 200.100.100.0 global cluster_interconnect
[root@db1 bin] #. / oifcfg delif-global eth0
[root@db1 bin] #. / oifcfg setif-global eth0/192.168.1.0:public
[root@db1 bin] #. / oifcfg getif
Eth2 200.100.100.0 global cluster_interconnect
Eth0 192.168.1.0 global public
[root@db1 bin] #. / oifcfg delif-global eth2
PRIF-31: Failed to delete the specified network interface because it is the last private interface
Deleting the last private interface is not allowed here. You can add a private interface first and then delete it.
[root@db1 bin] #. / oifcfg setif-global eth2/200.100.100.0:cluster_interconnect
[root@db1 bin] #. / oifcfg getif
Eth2 100.100.100.0 global cluster_interconnect
Eth0 192.168.2.0 global public
Eth2 200.100.100.0 global cluster_interconnect
[root@db1 bin] #. / oifcfg delif-global eth2/100.100.100.0:cluster_interconnect
[root@db1 bin] #. / oifcfg getif
Eth0 192.168.2.0 global public
Eth2 200.100.100.0 global cluster_interconnect
Stop the oracle high-availability service at a node
[root@db1 bin] #. / crsctl stop cluster-all
5. Modify the IP address and / etc/hosts file and test
6. Restart all services on all nodes
[root@db1 bin] #. / crsctl start cluster-all
7. Stop scan_listener and scan
[root@db1 bin] #. / srvctl stop scan_listener
[root@db1 bin] #. / srvctl stop scan
8. Delete scan_listener and scan
[root@db1 bin] #. / srvctl remove scan_listener-f
[root@db1 bin] #. / srvctl remove scan-f
9. Add scan and scan_listener
[root@db1 bin] #. / srvctl add scan-n db-scan-k 2-S 192.168.1.0/255.255.255.0/eth0
Description: an error will be reported if-k is 1 (network number) in the above command, because it has been used before
[root@db1 bin] #. / srvctl config scan
SCAN name: db-scan, Network: 2/192.168.1.0/255.255.255.0/eth0
SCAN VIP name: scan1, IP: / db-scan/192.168.1.75
[root@db1 bin] #. / srvctl add scan_listener
10. Start scan and scan_listener
[root@db1 bin] #. / srvctl start scan
[root@db1 bin] #. / srvctl start scan_listener
11. Stop VIP resources
[root@db1 bin] #. / crsctl stop resource ora.db1.vip-f
[root@db1 bin] #. / crsctl stop resource ora.db2.vip-f
[root@db1 bin] #. / crsctl stop resource ora.db3.vip-f
You can view the status of vip here with the following command
[root@db1 bin] #. / crsctl stat res-t
Ora.db1.vip
1 OFFLINE OFFLINE
Ora.db2.vip
1 OFFLINE OFFLINE
Ora.db3.vip
1 OFFLINE OFFLINE
12. Modify vip address
[root@db1 bin] #. / srvctl modify nodeapps-A 192.168.1.73/255.255.255.0/eth0-n db1
[root@db1 bin] #. / srvctl modify nodeapps-A 192.168.1.74/255.255.255.0/eth0-n db2
[root@db1 bin] #. / srvctl modify nodeapps-A 192.168.1.77/255.255.255.0/eth0-n db3
13. Start the VIP resource:
[root@db1 bin] #. / crsctl start resource ora.db1.vip-f
CRS-2672: Attempting to start 'ora.db1.vip' on' db1'
CRS-2676: Start of 'ora.db1.vip' on' db1' succeeded
[root@db1 bin] #. / crsctl start resource ora.db2.vip-f
CRS-2672: Attempting to start 'ora.db2.vip' on' db2'
CRS-2676: Start of 'ora.db2.vip' on' db2' succeeded
[root@db1 bin] #. / crsctl start resource ora.db3.vip-f
CRS-2672: Attempting to start 'ora.db3.vip' on' db3'
CRS-2676: Start of 'ora.db3.vip' on' db3' succeeded
After the VIP resource is started, local listening has already been started. The following step 14 may not be performed.
14. Start local monitoring:
[root@db1 bin] #. / srvctl start listener-n db1
[root@db1 bin] #. / srvctl start listener-n db2
[root@db1 bin] #. / srvctl start listener-n db3
Problem: execute oifcfg getif after the operation is complete, which appears in the prompt below
[grid@db1 peer] $oifcfg getif
Eth0 192.168.1.0 global public
Eth2 100.100.100.0 global cluster_interconnect
Only in OCR: eth2: 100.100.100.0 global cluster_interconnect
PRIF-30: Network information in OCR and GPnP profile differs
It is said that the network information is inconsistent between OCR and GPnP. You can reconfigure it here.
[grid@db1 peer] $oifcfg setif-global eth2/100.100.100.0:cluster_interconnect
[grid@db1 peer] $oifcfg getif
Eth0 192.168.1.0 global public
Eth2 100.100.100.0 global cluster_interconnect
