Tuesday, 2 August 2011

Fast Mirror resync and Fast Rebalance - ASM

     Fast Mirror resync and Fast Rebalance are  two new features available in 11g .

 Fast Mirror resync:

   This feature is usefull when
             1. Disk error is transient and it can be recovered easily within short  time ( short time means how long???.i would like to discuss about this in my next post)
             2. Temporary maintenance of a disk.
  ASM fast resync keeps track of pending changes to extents on an OFFLINE disk during an outage. The extents are resynced when the disk is brought back online or replaced.

By default, ASM drops a disk shortly after it is taken offline. You can set the DISK_REPAIR_TIME attribute to prevent this operation by specifying a time interval to repair the disk and bring it back online.

The default value for  DISK_REPAIR_TIME  is 3.6H


1) This feature requires that the redundancy level for the disk should be set to NORMAL or HIGH.

2) To use this feature, the disk group compatibility attributes must be set to 11.1 or higher
 compatible.asm & compatible.rdbms = 11.1.0.0.0 or higher

3) ALTER DISKGROUP dgroupA SET ATTRIBUTE 'DISK_REPAIR_TIME'='3H';

Fast Rebalance

Sometime DBAs dont want to do the expensive rebalance operation during the production hour after replacing (hot)  a disk. They may have downtime during week-end to do the rebalance.ofcourse fast rebalance is the best option for this

In 11g, you can use the STARTUP RESTRICT command to control access to an ASM instance while you perform maintenance. This startup option is used to implement ASM Fast Rebalance feature. Rebalance operations that occur while a disk group is in RESTRICTED mode eliminate the lock and unlock extent map messaging between ASM instances in Oracle RAC environments.

When an ASM instance is started with 'STARTUP RESTRICT' option and is active in this mode, all of the disk groups are mounted in RESTRICTED mode which prevents databases from connecting to the ASM instance. In addition, the restricted clause of the ALTER SYSTEM statement is disabled for the ASM instance.
  Follow the below steps to do the fast rebalance
  1. startup restrict;
  2.  alter diskgroup dg1 mount restricted;
            ( we cannot mount the disk in mount mode and may get ORA-15251: only restricted mount is allowed in the ASM instance )

3.alter diskgroup dg1 rebalance;
  ( we cannot open database against the asm when it is in restrict mode,we may get the below error


ORA-01157: cannot identify/lock data file 5 - see DBWR trace file
ORA-01110: data file 5: '+DG1/11g/datafile/exy.256.627345731'


4.shutdown immediate
   ( we cannot enable restriced mode as we do in normal rdbms
     we may the below error
ORA-15000: command disallowed by current instance type
)

5.startup

Alternatively we can dismount and mount in restricted mode a particular diskgroup when asm instance is running in non- restricted mode. This will help us when we have multiple diskgroups and each diskgroup is used by different database and other databases cannot afford downtime when fastrebalnce is happening.still if a diskgroup is mounted in restricted mode, no asm client can use that particular diskgroup

SQL> ALTER DISKGROUP data DISMOUNT;

Diskgroup altered.

SQL> ALTER DISKGROUP data MOUNT RESTRICTED;

Diskgroup altered.

SQL> ALTER DISKGROUP data DISMOUNT;

Diskgroup altered.

ASQL> ALTER DISKGROUP data MOUNT;

Diskgroup altered.

No comments:

Post a Comment

ZFS

Public Cloud tools comparison