Performing block-level recovery with Oracle Recovery Manager (RMAN)

Contributors NetAppZacharyWambold

You can configure SnapManager to catalog its backups in Recovery Manager (RMAN), an Oracle tool, so that you can perform a block-level recovery using RMAN. RMAN can use either the database’s control files or a separate recovery catalog database as its repository.

  1. To perform a full offline backup using SnapManager, enter the following command:

    smo backup create -offline-full-profileprofile_name-labelbackup_label_name-verbose

    Where:

    • profile_name is the name of the profile associated with the backup

    • backup_label_name is the name of the backup label

smo backup create -offline -full -profile profile_monthly
-label full_backup -verbose

+
SMO-07109 [INFO ]: Cataloguing all files in backup set with RMAN
TAG=SMC_full_backup_1158773581857, RMAN=ES0/controlfile.
...
SMO-13037 [INFO ]: Successfully completed operation: Backup
SMO-13048 [INFO ]: Operation Status: SUCCESS
SMO-13049 [INFO ]: Elapsed Time: 0:02:20.506
Operation Id [ff8080810dcc47e3010dcc47eb7a0001] succeeded.
+

  1. To verify that the backup is cataloged with RMAN, from the database host, enter the following command at the RMAN prompt:

    list datafilecopy tag tag_name;

  2. To verify the database and determine if any blocks are corrupted, enter the following command:

    dbv FILE=user01.dbf

    The following output shows that two pages are corrupt:

    DBVERIFY: Release 10.2.0.1.0 - Production on Wed Sep 20 13:35:44 2006
Copyright (c) 1982, 2005, Oracle. All rights reserved.
DBVERIFY - Verification starting : FILE = user01.dbf
Page 625 is marked corrupt
Corrupt block relative dba: 0x01400271 (file 5, block 625)
Bad header found during dbv:
Data in bad block:
type: 240 format: 6 rdba: 0xed323b81
last change scn: 0x6f07.faa74628 seq: 0x87 flg: 0x02
spare1: 0x60 spare2: 0x5 spare3: 0xef7d
consistency value in tail: 0xa210fe71
check value in block header: 0x13c7
block checksum disabled...
Page 627 is marked corrupt
Corrupt block relative dba: 0x01400273 (file 5, block 627)
Bad header found during dbv:
Data in bad block:
type: 158 format: 7 rdba: 0x2101e16d
last change scn: 0xe828.42414628 seq: 0xb4 flg: 0xff
spare1: 0xcc spare2: 0x81 spare3: 0x8665
consistency value in tail: 0x46d20601
check value in block header: 0x1a84
computed block checksum: 0x6c30
DBVERIFY - Verification complete
Total Pages Examined : 1280
Total Pages Processed (Data) : 1123
Total Pages Failing (Data) : 0
Total Pages Processed (Index): 0
Total Pages Failing (Index): 0
Total Pages Processed (Other): 34
Total Pages Processed (Seg) : 0
Total Pages Failing (Seg) : 0
Total Pages Empty : 120
Total Pages Marked Corrupt: 2
Total Pages Influx : 0
Highest block SCN : 1337349 (0.1337349)

  3. To make the files from the backup accessible on the host and to RMAN, mount the backup by using the following command:

    smo backup mount -profileprofile_name-labellabel-verbose

    smo backup mount -profile SALES1 -label full_backup -verbose

SMO-13046 [INFO ]: Operation GUID 8abc013111b9088e0111b908a7560001 starting on Profile SALES1
SMO-08052 [INFO ]: Beginning to connect mount(s) [E:\logs,F:\data] from logical snapshot SMO_SALES1_hsdb1_F_C_1_8abc013111a450480111a45066210001.
SMO-08025 [INFO ]: Beginning to connect mount E:\logs from snapshot SMO_SALES1_hsdb1_F_C_1_8abc013111a450480111a45066210001_0 of volume hs_logs.
SMO-08027 [INFO ]: Finished connecting mount E:\logs from snapshot SMO_SALES1_hsdb1_F_C_1_8abc013111a450480111a45066210001_0 of volume hs_logs.
SMO-08025 [INFO ]: Beginning to connect mount F:\data from snapshot SMO_SALES1_hsdb1_F_C_1_8abc013111a450480111a45066210001_0 of volume hs_data.
SMO-08027 [INFO ]: Finished connecting mount F:\data from snapshot SMO_SALES1_hsdb1_F_C_1_8abc013111a450480111a45066210001_0 of volume hs_data.
SMO-08053 [INFO ]: Finished connecting mount(s) [E:\logs,F:\data] from logical snapshot SMO_SALES1_hsdb1_F_C_1_8abc013111a450480111a45066210001.
SMO-13037 [INFO ]: Successfully completed operation: Backup Mount
SMO-13048 [INFO ]: Operation Status: SUCCESS
SMO-13049 [INFO ]: Elapsed Time: 0:01:00.981
Operation Id [8abc013111b9088e0111b908a7560001] succeeded.

  4. To recover the blocks, in RMAN, enter the following command:

    blockrecover datafile ‘E:\path\file.dbf’ block block_id, from tag backup_rman_tag

    RMAN> blockrecover datafile
‘E:\sys\file01.dbf’ block 625, 626, 627
from tag SMO_full_backup_1158773581857;

Starting blockrecover at 20-SEP-08 using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: sid=153 devtype=DISK
channel ORA_DISK_1: restoring block(s) from datafile copy
 C:\myfs\user01.dbf
starting media recovery
media recovery complete, elapsed time: 00:00:01
Finished blockrecover at 20-SEP-08

  5. To verify if the blocks have been repaired, use the following command:

    dbv FILE=filename.dbf

    The following output shows that no pages are corrupt:

    dbv FILE=user01.dbf

DBVERIFY: Release 10.2.0.1.0 - Production on Wed Sep 20 13:40:01 2008
Copyright (c) 1982, 2008, Oracle. All rights reserved.
DBVERIFY - Verification starting : FILE = user01.dbf
DBVERIFY - Verification complete
Total Pages Examined : 1280
Total Pages Processed (Data) : 1126
Total Pages Failing (Data) : 0
Total Pages Processed (Index): 0
Total Pages Failing (Index): 0
Total Pages Processed (Other): 34
Total Pages Processed (Seg) : 0
Total Pages Failing (Seg) : 0
Total Pages Empty : 120
Total Pages Marked Corrupt : 0
Total Pages Influx : 0
Highest block SCN : 1337349 (0.1337349)

    All corrupted blocks were repaired and restored.