rajesh: Rajesh's Weblog

 I try to summarize here some of the errors and its solutions i got while running a Oracle 10g R2 Real Application Cluster (RAC) Install on VMware Linux install.

starting VIP application resource on (2) nodes1:CRS-1006: No more members to consider
CRS-0215: Could not start resours 'ora.vm01rhel.vip'  : Check the log file /u01/app/oracle/product/10.2.0/crs/log/vm01rhel/racg/ora.vm01rhel.vip.log
- I got this issue resolved by specifying the default gateway to some dummy ip address (same subnet) in both of my virtual machines. Earlier i didnt specify any default gateway both in my virtual machines and got this error

CLUSTERWARE ISSUES: 
Oracle Notification Server Configuration Assistant (Clusterware) - FAIL
(http://forums.oracle.com/forums/message.jspa?messageID=1225245)

 
OCFS2 ISSUES: 
http://www.rampant-books.com/art_hunter_rac_oracle_o2cb_cluster_service.htm
I got some ocfs2 issues due to the wrong rpm i selected
First do uname -a and then see which OS you have got ,rpm's which supports the same version needs to be downloaded from
http://oss.oracle.com/projects/ocfs2-tools/files/RedHat/RHEL4/i386/1.2.1-1/
http://oss.oracle.com/projects/ocfs2/files/

While installing Oracle Clusterware and database i got an error "Remote 'AttachHome' failed on nodes:"
After installation run this command as a Oracle user and attach home will be successfull
/u01/app/oracle/product/10.2.0/db_1/oui/bin/runInstaller -attachHome -noClusterEnabled ORACLE_HOME=/u01/app/oracle/product/10.2.0/db_1 ORACLE_HOME_NAME=OraDb10g_home1
CLUSTER_NODES=vm01rhel,vm02rhel "INVENTORY_LOCATION=/u01/app/oracle/oraInventory" LOCAL_NODE=vm01rhel(or vm02rhel)
Note : In case you are running this command after a clusterware installation , replace 'db_1' with 'crs'

How do i Remove Oracle Database and Clusterware thru command line (incase the oui doesnt uninstall it)
/u01/app/oracle/product/10.2.0/db_1/oui/bin/runInstaller -removeHome -noClusterEnabled ORACLE_HOME=/u01/app/oracle/product/10.2.0/db_1 ORACLE_HOME_NAME=OraDb10g_home1 CLUSTER_NODES=vm01rhel,vm02rhel "INVENTORY_LOCATION=/u01/app/oracle/oraInventory" LOCAL_NODE=vm01rhel(or vm02rhel)
-where vm01rhel / vm02rhel are your node names
-OraDb10g_home1 is the Oracle DB Home name

Note : In case you are running this command after a clusterware installation , replace 'db_1' with 'crs'

Running root.sh fails (after installing clusterware and database) OR ocrconfig error
Disk volume needs to be mounted as "datavolume" -> mount -t ocfs -o datavolume /dev/sdc1 /u02/../ocfs)
Edit /etc/fstab and add this line to mount the disks (sdc1, sdd1 etc) permanantly mountable
/dev/sdc1 /ocfs ocfs _netdev,datavolume 0 0
/dev/sdd1 /ocfs ocfs _netdev,datavolume 0 0

INITPROT : Failed to start cluster
-> unmount the shared disk/partition (sdc1, sdd1 etc)
-> o2cb offline/unload/ load/start/ cluster
-> now run root.sh again (sometimes you may need to delete the /etc/oracle directory to re run this command successfully)

Oracle Server Notifcation Failed: 
http://forums.oracle.com/forums/message.jspa?messageID=1225245

After Restarting of RAC Database instance machines, the DB admin console(http://:1158/em) is not accessible: After Restarting the nodes, Login as Oracle user
- Now start the emctl (Enterprise Manager) dbconsole . This will start the dbconsole. Now you can access the console and do the db operations by accessing http://:1158/em
$emctl start

Cannot shutdown the Database instance from DBconsole:  Do this for all the nodes. Now try to shutdown the Db instance
- $srvctl stop nodeapps