A component called “OLR” is introduced in 11gR2, which will be installed on every node of Cluster. This component is not shared by other nodes and is specific to the respective node. OLR will be setup when Oracle Clusterware is installed and configured.
As we all know OCR will be accessed first, to find the resources that need to be started on the respective node, in the startup phase. But, if we use ASM as the Storage Management, and store OCR and Voting disk on ASM, the OCR will not be accessed until ASM is started. So, OLR plays a key role in the identifying services that need to be started; once the services (like ASM) are started, OCR can be accessed.
OLR is an Operating System File, and is located in $GRID_HOME/cdata/<hostname>.olr. The location of OLR is stored in /etc/oracle/olr.loc and is used by OHASD (Oracle High Availabilitty Services Daemon).
Important: If OLR is missing or corrupted, Oracle Clusterware cannot be started on the respective node.
Some important commands:
- Check the status of OLR file on each node by “ocrcheck -local”
- Output the contents of OLR to text file using “ocrdump -local -stdout”
- Export/import OLR with commands, “ocrconfig -local -export <file name>”, “ocrconfig -local -import <filename>”
- Repair OLR using backup, “ocrconfig -local -repair olr <filename>
- Manual backup, ocrconfig -local -manualbackup