Uninstall & Reinstall RAC (10GR2) on Solaris 10 SPARC

Question

Hi,
I'm trying to install 2 node RAC. I got below errors while executing "root.sh" at the end of installation on both nodes.
[client(19083)]CRS-1011:OCR cannot determine that the OCR content contains the latest updates.
Oracle Database 10g CRS Release 10.2.0.1.0 Production Copyright 1996, 2005 Oracle. All rights reserved.
2011-06-09 18:37:58.698: [ OCRCONF][1]ocrconfig starts...
2011-06-09 18:37:58.701: [ OCRCONF][1]Upgrading OCR data
2011-06-09 18:37:59.363: [ OCRRAW][1]propriogid:1: INVALID FORMAT
2011-06-09 18:37:59.363: [ OCRRAW][1]proprioini: disk 0 (/dev/rdsk/emcpower5e) doesn't have enough votes (1,2)
2011-06-09 18:37:59.376: [ OCRRAW][1]proprinit: Could not open raw device
2011-06-09 18:37:59.376: [ default][1]a_init:7!: Backend init unsuccessful : [26]
2011-06-09 18:37:59.377: [ OCRCONF][1]Exporting OCR data to [OCRUPGRADEFILE]
2011-06-09 18:37:59.377: [ OCRAPI][1]a_init:7!: Backend init unsuccessful : [33]
2011-06-09 18:37:59.377: [ OCRCONF][1]There was no previous version of OCR. error:[PROC-33: Oracle Cluster Registry is not configured]
2011-06-09 18:37:59.393: [ OCRRAW][1]propriogid:1: INVALID FORMAT
2011-06-09 18:37:59.393: [ OCRRAW][1]proprioini: disk 0 (/dev/rdsk/emcpower5e) doesn't have enough votes (1,2)
2011-06-09 18:37:59.394: [ OCRRAW][1]proprinit: Could not open raw device
2011-06-09 18:37:59.394: [ default][1]a_init:7!: Backend init unsuccessful : [26]
2011-06-09 18:37:59.409: [ OCRRAW][1]propriogid:1: INVALID FORMAT
2011-06-09 18:37:59.409: [ OCRRAW][1]proprioini: disk 0 (/dev/rdsk/emcpower5e) doesn't have enough votes (1,2)
2011-06-09 18:37:59.410: [ OCRRAW][1]proprinit: Could not open raw device
2011-06-09 18:37:59.410: [ OCRAPI][1]a_init:6b!: Backend init unsuccessful : [26]
2011-06-09 18:37:59.410: [ OCRCONF][1]Failed to initialized OCR context. error:[PROC-26: Error while accessing the physical storage]
2011-06-09 18:37:59.411: [ OCRCONF][1]Exiting [status=failed]...
The permissions for OCR disks on both the nodes were "root:dba" and the disks was accessible across the nodes.
I'm not sure why we got that error.
To over come it, I'm going as below:
- rootdelete.sh on both nodes
- rootdeinstall on node 1
- uninstall using OUI
- remove /var/opt/oracle/*
- remove oracle_home
- remove oraInventory
- format the disks
- reboot the nodes.
Just wanted to know if there's something else that should be done.
thanks

Answer 1

I strongly advise against uninstalling Clusterware without first making a serious attempt to troubleshoot what went wrong and trying to fix the errors. If you unistall and retry without knowing what to do differently, you will have hardly any choice but to repeat the same steps that led to the errors in root.sh
>
The permissions for OCR disks on both the nodes were "root:dba" and the disks was accessible across the nodes.
>
That could be your error right there. The installation instructions say to put the voting disk in the dba group, but to put the cluster registry in the oinstall group. Quoting from page 4-7 of the Clusterware and RAC installation guide:
http://download.oracle.com/docs/cd/B19306_01/install.102/b14205.pdf
>
– The voting disk is a partition that Oracle Clusterware uses to verify cluster
node membership and status.
The voting disk must be owned by the oracle user, must be in the dba group,
and must have permissions set to 644. Provide at least 20 MB disk space for
the voting disk.
– The Oracle Cluster Registry (OCR) contains cluster and database configuration
information for the RAC database and for Oracle Clusterware, including the
node list, and other information about cluster configuration and profiles.
The OCR disk must be owned by root, must be in the oinstall group, and
must have permissions set to 640. Provide at least 100 MB disk space for the
OCR.