咨询微信: dbservice1234 7 x 24 在线支持!

    你在这里

    • You are here:
    • 首页 > 博客 > PDSERVICE的博客 > How To Restore/Repair/Fix An Overwritten (KFBTYP_INVALID) Oracle ASM Disk Header (First 4K) 10.2.0.5, 11.1.0.7, 11.2 And Onwards

How To Restore/Repair/Fix An Overwritten (KFBTYP_INVALID) Oracle ASM Disk Header (First 4K) 10.2.0.5, 11.1.0.7, 11.2 And Onwards

How To Restore/Repair/Fix An Overwritten (KFBTYP_INVALID) Oracle ASM Disk Header (First 4K) 10.2.0.5, 11.1.0.7, 11.2 And Onwards

APPLIES TO:
Oracle Database Cloud Schema Service - Version N/A and later
Oracle Database Exadata Cloud Machine - Version N/A and later
Oracle Database Exadata Express Cloud Service - Version N/A and later
Oracle Cloud Infrastructure - Database Service - Version N/A and later
Oracle Database Backup Service - Version N/A and later
Information in this document applies to any platform.
GOAL
The present document provides an example about “how to restore/repair/fix an overwritten ASM Disk Header (first 4K) on 11.1.0.7 and Onwards”.
 
SOLUTION
A copy of the ASM disk header (first 4K) exists on 10.2.0.5, 11.1.0.7, 11.2 and onwards. It can be used to try to restore a valid ASM disk header (Assuming only the first 4k of the disk were affected/overwritten). In order to restore the ASM disk header (assuming the automatic ASM disk header backup is in good shape) please perform the next steps:
 
 
1) Backup the first 50MB of the affected disk (this step is mandatory):
 
$> dd if=<full path affected disk name> of=/tmp/<affected disk name>.dump bs=1048576 count=50
 
Example:
 
[grid@dbaasm ~]$ dd if=/dev/oracleasm/disks/ASMDISK2 of=/tmp/ASMDISK2.dump bs=1048576 count=50
50+0 records in
50+0 records out
52428800 bytes (52 MB) copied, 0.667474 seconds, 78.5 MB/s
 
 
Where: "/dev/oracleasm/disks/ASMDISK2" is the affected ASM disk member .
 
2)  Collect the Allocation Unit Size (kfdhdb.ausize) from another healthy disk member (from the same affected diskgroup):
 
$> <ASM Oracle Home>/bin/kfed read <full path healthy disk name> | egrep 'ausize|dsknum|dskname|grpname|fgname'  
 
Example:
 
[grid@dbaasm ~]$ kfed read /dev/oracleasm/disks/ASMDISK1  | egrep 'ausize|dsknum|dskname|grpname|fgname'  
 
kfdhdb.dsknum:                        0 ; 0x024: 0x0000
kfdhdb.dskname:                ASMDISK1 ; 0x028: length=8
kfdhdb.grpname:                    DATA ; 0x048: length=4
kfdhdb.fgname:                 FG1_SAN1 ; 0x068: length=8
kfdhdb.ausize:                  2097152 ; 0x0bc: 0x00200000
 
 
Note: In this example, the diskgroup was created using an AU_SIZE=2M (2097152 ) & "/dev/oracleasm/disks/ASMDISK1" is the healthy ASM disk member .
 
3) Then restore the ASM disk header from backup as follows:
 
$> <ASM Oracle Home>/bin/kfed repair <full path affected disk name> ausz=<AU size from point #2>
 
Example:
 
[grid@dbaasm ~]$ kfed repair /dev/oracleasm/disks/ASMDISK2  ausz=2097152
 
4) Verify that the ASM disk header in the affected disk was recreated/restored:
 
$> <ASM Oracle Home>/bin/kfed read <full path affected disk name>  | head -40
 
Example:
 
[grid@dbaasm ~]$  kfed read /dev/oracleasm/disks/ASMDISK2  | head -40
kfbh.endian:                          1 ; 0x000: 0x01
kfbh.hard:                          130 ; 0x001: 0x82
kfbh.type:                            1 ; 0x002: KFBTYP_DISKHEAD
kfbh.datfmt:                          1 ; 0x003: 0x01
kfbh.block.blk:                       0 ; 0x004: blk=0
kfbh.block.obj:              2147483650 ; 0x008: disk=2
kfbh.check:                  4052202307 ; 0x00c: 0xf187b343
kfbh.fcn.base:                        0 ; 0x010: 0x00000000
kfbh.fcn.wrap:                        0 ; 0x014: 0x00000000
kfbh.spare1:                          0 ; 0x018: 0x00000000
kfbh.spare2:                          0 ; 0x01c: 0x00000000
kfdhdb.driver.provstr: ORCLDISKASMDISK2 ; 0x000: length=16
kfdhdb.driver.reserved[0]:   1145918273 ; 0x008: 0x444d5341
kfdhdb.driver.reserved[1]:    843797321 ; 0x00c: 0x324b5349
kfdhdb.driver.reserved[2]:            0 ; 0x010: 0x00000000
kfdhdb.driver.reserved[3]:            0 ; 0x014: 0x00000000
kfdhdb.driver.reserved[4]:            0 ; 0x018: 0x00000000
kfdhdb.driver.reserved[5]:            0 ; 0x01c: 0x00000000
kfdhdb.compat:                186647296 ; 0x020: 0x0b200300
kfdhdb.dsknum:                        2 ; 0x024: 0x0002
kfdhdb.grptyp:                        2 ; 0x026: KFDGTP_NORMAL
kfdhdb.hdrsts:                        3 ; 0x027: KFDHDR_MEMBER
kfdhdb.dskname:                ASMDISK2 ; 0x028: length=8
kfdhdb.grpname:                    DATA ; 0x048: length=4
kfdhdb.fgname:                 FG2_SAN2 ; 0x068: length=8
kfdhdb.capname:                         ; 0x088: length=0
kfdhdb.crestmp.hi:             32974423 ; 0x0a8: HOUR=0x17 DAYS=0x12 MNTH=0x9 YEAR=0x7dc
kfdhdb.crestmp.lo:           1180930048 ; 0x0ac: USEC=0x0 MSEC=0xe4 SECS=0x26 MINS=0x11
kfdhdb.mntstmp.hi:             33003184 ; 0x0b0: HOUR=0x10 DAYS=0x15 MNTH=0x5 YEAR=0x7de
kfdhdb.mntstmp.lo:           1230240768 ; 0x0b4: USEC=0x0 MSEC=0xff SECS=0x15 MINS=0x12
kfdhdb.secsize:                     512 ; 0x0b8: 0x0200
kfdhdb.blksize:                    4096 ; 0x0ba: 0x1000
kfdhdb.ausize:                  2097152 ; 0x0bc: 0x00200000
kfdhdb.mfact:                    228480 ; 0x0c0: 0x00037c80
kfdhdb.dsksize:                    9769 ; 0x0c4: 0x00002629
kfdhdb.pmcnt:                         2 ; 0x0c8: 0x00000002
kfdhdb.fstlocn:                       1 ; 0x0cc: 0x00000001
kfdhdb.altlocn:                       2 ; 0x0d0: 0x00000002
kfdhdb.f1b1locn:                      0 ; 0x0d4: 0x00000000
kfdhdb.redomirrors[0]:                0 ; 0x0d8: 0x0000
.
.
.
.
 
5) Finally, mount the diskgroup:
 
SQL> alter diskgroup <diskgroup name> mount ;
 
Example:
 
 
SQL> alter diskgroup DATA mount;
 
Diskgroup altered.
 
 
Notes
Note 1: The solution provided in this document will work if the following conditions are true:
 
   a) Only the first 4K of the affected disk were overwritten/wiped out/overlapped.
 
   b) ASM disk header backup is in good shape.
 
 
 
Note 2: If this solution does not solve your problem, then do not attempt additional steps/actions on the affected diskgroup, therefore please engage Oracle Support through a new Service Request to determinate the “Root Cause” & possible solutions.
 
 
 
Note 3: An ASM disk with a corrupted “Disk Header” will report the following output: 
 
[grid@dbaasm ~]$ kfed read /dev/oracleasm/disks/ASMDISK2
 
kfbh.endian:                          0 ; 0x000: 0x00
 
kfbh.hard:                            0 ; 0x001: 0x00
 
kfbh.type:                            0 ; 0x002: KFBTYP_INVALID
 
kfbh.datfmt:                          0 ; 0x003: 0x00
 
kfbh.block.blk:                       0 ; 0x004: blk=0
 
kfbh.block.obj:                       0 ; 0x008: file=0
 
kfbh.check:                           0 ; 0x00c: 0x00000000
 
kfbh.fcn.base:                        0 ; 0x010: 0x00000000
 
kfbh.fcn.wrap:                        0 ; 0x014: 0x00000000
 
kfbh.spare1:                          0 ; 0x018: 0x00000000
 
kfbh.spare2:                          0 ; 0x01c: 0x00000000
 
000000000 00000000 00000000 00000000 00000000  [................]
 
  Repeat 255 times
 
KFED-00322: Invalid content encountered during block traversal: [kfbtTraverseBlock][Invalid OSM block type][][0]