Archive for May 2011

How to Store Voting Disk Files in ASM for 11g R2 Clusterware High Availability   2 comments

11gR2 clusterware voting disk files can be stored in ASM through the ASM diskgroup.  To ensure the high availability of the clusterware, it is highly recommended to store multiple voting disk files. When storing the voting disk files in ASM, the redundancy setting of the ASM diskgroup determines the number of voting disk files. The following settings  are recommended for planning  the proper configuration of ASM diskgroup that is used to store the voting disk files.

  • External Redundancy: The ASM diskgroup only needs one failure group without mirroring and it only stores one voting disk file. It is recommended that the disk have  external redundancy RAID configuration to provide the storage high availability.  
  • Normal Redundancy:  The ASM diskgroup requires three failure groups and it will provide three voting disks files.
  • High Redundancy: The ASM diskgroup requires five failure groups and it has five voting disk files.

Following the recommended settings above, you can configure the 11g R2 clusterware by storing the voting disk files in ASM:

1.  Based on the external storage redundancy and number of voting disks that you plan  to have , select the appropriate redundancy setting for your ASM diskgroup. If there is no external redundancy for the storage disks or you would like to have additional voting disk files, it is recommended to select normal or high redundancy as the setting of choice for the ASM diskgroup.  As an example , we configure the high redundancy setting for the ASM diskgroup. The ASM diskgroup has five ASM failure groups each of which is on one of the following ASM disks:  ORCL:OCR1,  ORCL: OCR2, ORCL:OCR3,  ORCL: OCR4, ORCL:OCR5

     2.   During 11g R2 Oracle Grid Infrastructure installation,  select “Automatic Storage Management (ASM) In the storage selection step  as shown below:

3. Specify the ASM diskgroup name and select the proper redundancy level and the corresponding ASM disks. As   shown in the example below, select the high redundancy and the five ASM disk: ORCL:OCR1,  ORCL: OCR2, ORCL:OCR3,  ORCL: OCR4, ORCL:OCR5

4. After the Grid Infrastructure installation completes, login to the ASM instance to check the ASM diskgroup redundancy setting and list the voting disk files using the command of ‘crsctl query css votedisk’.

My Upcoming Presentation on IOUG –RAC Webinar Series   Leave a comment

I will speak at IOUG RAC Webinar Series on May 11th, 2011. The presentation aims to provide DBAs some practical understanding and best practices of system configuration such as network and storage that are very critical to the stability of Oracle clusterware and RAC. It discusses hardware/OS configuration, public and private interconnect network and shared storage configuration, and shares some experiences/tips for troubleshooting clusterware issues such as node eviction and how some useful diagnostic tools help for the root cause analysis. This presentation also covers some clusterware administration methods as well as the new clusterware features. The following are the details of this IOUG Webinar:

Title: Oracle11g R2 Clusterware: Architecture and Best Practices of Configuration and Troubleshooting
Date:
May 11, 2011 from 11:00am-12:00pm CDT
Register Today! https://www1.gotomeeting.com/register/833030457  

Outlines:

  • Oracle 11g R2 Clusterware Architecture
  • System Hardware for Oracle Clusterware/RAC
  • Network configuration for Oracle Clusterware/RAC
  • Storage Configuration for Oracle Clusterware/RAC
  • Managing Oracle Clusterware
  • Clusterware Troubleshooting
  • QA