标签归档:endian_kfbh

ORA-15196: invalid ASM block header [kfc.c:26368] [endian_kfbh]故障处理

客户对asm进行扩容,由于配置不恰当,在使用asmca增加asm disk的时候直接选中了已经被用作文件系统的vg中的磁盘

Tue Nov 19 09:48:48 2019
Non critical error ORA-48180 cFri Nov 22 12:47:48 2019
SQL> ALTER DISKGROUP XIFENFEI ADD  DISK '/dev/rhdisk29' SIZE 491520M ,
'/dev/rhdisk30' SIZE 491520M ,
'/dev/rhdisk31' SIZE 491520M /* ASMCA */ 
NOTE: GroupBlock outside rolling migration privileged region
NOTE: Assigning number (4,15) to disk (/dev/rhdisk29)
NOTE: Assigning number (4,16) to disk (/dev/rhdisk30)
NOTE: Assigning number (4,17) to disk (/dev/rhdisk31)
NOTE: requesting all-instance membership refresh for group=4
NOTE: initializing header on grp 4 disk XIFENFEI_0015
NOTE: initializing header on grp 4 disk XIFENFEI_0016
NOTE: initializing header on grp 4 disk XIFENFEI_0017
NOTE: requesting all-instance disk validation for group=4
Fri Nov 22 12:47:51 2019
NOTE: skipping rediscovery for group 4/0xb08c40b (XIFENFEI) on local instance.
NOTE: requesting all-instance disk validation for group=4
NOTE: skipping rediscovery for group 4/0xb08c40b (XIFENFEI) on local instance.
Fri Nov 22 12:47:59 2019
NOTE: initiating PST update: grp = 4
Fri Nov 22 12:47:59 2019
GMON updating group 4 at 12 for pid 27, osid 12649908
NOTE: PST update grp = 4 completed successfully 
NOTE: membership refresh pending for group 4/0xb08c40b (XIFENFEI)
GMON querying group 4 at 13 for pid 18, osid 39912680
Fri Nov 22 12:48:01 2019
NOTE: cache opening disk 15 of grp 4: XIFENFEI_0015 path:/dev/rhdisk29
NOTE: cache opening disk 16 of grp 4: XIFENFEI_0016 path:/dev/rhdisk30
NOTE: cache opening disk 17 of grp 4: XIFENFEI_0017 path:/dev/rhdisk31
NOTE: Attempting voting file refresh on diskgroup XIFENFEI
NOTE: Refresh completed on diskgroup XIFENFEI. No voting file found.
GMON querying group 4 at 14 for pid 18, osid 39912680
SUCCESS: refreshed membership for 4/0xb08c40b (XIFENFEI)
SUCCESS: ALTER DISKGROUP XIFENFEI ADD  DISK '/dev/rhdisk29' SIZE 491520M ,
'/dev/rhdisk30' SIZE 491520M ,
'/dev/rhdisk31' SIZE 491520M /* ASMCA */

发现增加错磁盘之后,从vg里面强制踢掉被asm使用的磁盘,并且尝试在asm中删除这些磁盘,并加入新磁盘

Fri Nov 22 12:52:03 2019
SQL> ALTER DISKGROUP XIFENFEI DROP  DISK 'XIFENFEI_0015','XIFENFEI_0016','XIFENFEI_0017' /* ASMCA */ 
NOTE: GroupBlock outside rolling migration privileged region
Fri Nov 22 12:52:03 2019
NOTE: stopping process ARB0
NOTE: rebalance interrupted for group 4/0xb08c40b (XIFENFEI)
NOTE: requesting all-instance membership refresh for group=4
NOTE: membership refresh pending for group 4/0xb08c40b (XIFENFEI)
Fri Nov 22 12:52:12 2019
GMON querying group 4 at 15 for pid 18, osid 39912680
SUCCESS: refreshed membership for 4/0xb08c40b (XIFENFEI)
SUCCESS: ALTER DISKGROUP XIFENFEI DROP  DISK 'XIFENFEI_0015','XIFENFEI_0016','XIFENFEI_0017' /* ASMCA */
NOTE: starting rebalance of group 4/0xb08c40b (XIFENFEI) at power 1
Starting background process ARB0
…………
Fri Nov 22 12:58:26 2019
SQL> ALTER DISKGROUP XIFENFEI ADD  DISK '/dev/rhdisk7' SIZE 491520M /* ASMCA */ 
NOTE: GroupBlock outside rolling migration privileged region
Fri Nov 22 12:58:26 2019
NOTE: stopping process ARB0
NOTE: rebalance interrupted for group 4/0xb08c40b (XIFENFEI)
NOTE: ASM did background COD recovery for group 4/0xb08c40b (XIFENFEI)
NOTE: Assigning number (4,18) to disk (/dev/rhdisk7)
NOTE: requesting all-instance membership refresh for group=4
NOTE: initializing header on grp 4 disk XIFENFEI_0018
NOTE: requesting all-instance disk validation for group=4
NOTE: skipping rediscovery for group 4/0xb08c40b (XIFENFEI) on local instance.
NOTE: requesting all-instance disk validation for group=4
NOTE: skipping rediscovery for group 4/0xb08c40b (XIFENFEI) on local instance.
Fri Nov 22 12:58:41 2019
NOTE: initiating PST update: grp = 4
Fri Nov 22 12:58:41 2019
GMON updating group 4 at 16 for pid 27, osid 12649908
NOTE: PST update grp = 4 completed successfully 
Fri Nov 22 12:58:41 2019
NOTE: membership refresh pending for group 4/0xb08c40b (XIFENFEI)
GMON querying group 4 at 17 for pid 18, osid 39912680
NOTE: cache opening disk 18 of grp 4: XIFENFEI_0018 path:/dev/rhdisk7
NOTE: Attempting voting file refresh on diskgroup XIFENFEI
NOTE: Refresh completed on diskgroup XIFENFEI. No voting file found.
GMON querying group 4 at 18 for pid 18, osid 39912680
SUCCESS: refreshed membership for 4/0xb08c40b (XIFENFEI)
NOTE: starting rebalance of group 4/0xb08c40b (XIFENFEI) at power 1
SUCCESS: ALTER DISKGROUP XIFENFEI ADD  DISK '/dev/rhdisk7' SIZE 491520M /* ASMCA */
Starting background process ARB0
Fri Nov 22 12:58:46 2019
ARB0 started with pid=44, OS id=54460432 
…………
Fri Nov 22 12:59:57 2019
SQL> ALTER DISKGROUP XIFENFEI ADD  DISK '/dev/rhdisk10' SIZE 491520M ,
'/dev/rhdisk11' SIZE 491520M ,
'/dev/rhdisk8' SIZE 491520M ,
'/dev/rhdisk9' SIZE 491520M /* ASMCA */ 
NOTE: GroupBlock outside rolling migration privileged region
Fri Nov 22 12:59:57 2019
NOTE: stopping process ARB0
NOTE: rebalance interrupted for group 4/0xb08c40b (XIFENFEI)
NOTE: ASM did background COD recovery for group 4/0xb08c40b (XIFENFEI)
NOTE: Assigning number (4,19) to disk (/dev/rhdisk10)
NOTE: Assigning number (4,20) to disk (/dev/rhdisk11)
NOTE: Assigning number (4,21) to disk (/dev/rhdisk8)
NOTE: Assigning number (4,22) to disk (/dev/rhdisk9)
NOTE: requesting all-instance membership refresh for group=4
NOTE: initializing header on grp 4 disk XIFENFEI_0019
NOTE: initializing header on grp 4 disk XIFENFEI_0020
NOTE: initializing header on grp 4 disk XIFENFEI_0021
NOTE: initializing header on grp 4 disk XIFENFEI_0022
NOTE: requesting all-instance disk validation for group=4
NOTE: skipping rediscovery for group 4/0xb08c40b (XIFENFEI) on local instance.
Fri Nov 22 13:00:08 2019
NOTE: requesting all-instance disk validation for group=4
Fri Nov 22 13:00:08 2019
NOTE: skipping rediscovery for group 4/0xb08c40b (XIFENFEI) on local instance.
NOTE: initiating PST update: grp = 4
Fri Nov 22 13:00:13 2019
GMON updating group 4 at 19 for pid 27, osid 12649908
NOTE: PST update grp = 4 completed successfully 
NOTE: membership refresh pending for group 4/0xb08c40b (XIFENFEI)
GMON querying group 4 at 20 for pid 18, osid 39912680
NOTE: cache opening disk 19 of grp 4: XIFENFEI_0019 path:/dev/rhdisk10
NOTE: cache opening disk 20 of grp 4: XIFENFEI_0020 path:/dev/rhdisk11
NOTE: cache opening disk 21 of grp 4: XIFENFEI_0021 path:/dev/rhdisk8
NOTE: cache opening disk 22 of grp 4: XIFENFEI_0022 path:/dev/rhdisk9
NOTE: Attempting voting file refresh on diskgroup XIFENFEI
NOTE: Refresh completed on diskgroup XIFENFEI. No voting file found.
GMON querying group 4 at 21 for pid 18, osid 39912680
SUCCESS: refreshed membership for 4/0xb08c40b (XIFENFEI)
SUCCESS: ALTER DISKGROUP XIFENFEI ADD  DISK '/dev/rhdisk10' SIZE 491520M ,
'/dev/rhdisk11' SIZE 491520M ,
'/dev/rhdisk8' SIZE 491520M ,
'/dev/rhdisk9' SIZE 491520M /* ASMCA */
NOTE: starting rebalance of group 4/0xb08c40b (XIFENFEI) at power 1
Starting background process ARB0

asm在做着reblance的过程中遭遇到坏块,直接导致磁盘组dismount

Sun Nov 24 04:42:27 2019
NOTE: group 4 PST updated.
WARNING: cache read  a corrupt block: group=4(XIFENFEI) dsk=15 blk=258 disk=15
 (XIFENFEI_0015) incarn=1717056824 au=113792 blk=2 count=254
Errors in file /u01/app/oracle/diag/asm/+asm/+ASM2/trace/+ASM2_x000_28639240.trc:
ORA-15196: invalid ASM block header [kfc.c:26368] [endian_kfbh] [2147483663] [258] [56 != 0]
NOTE: a corrupted block from group XIFENFEI was dumped to
   /u01/app/oracle/diag/asm/+asm/+ASM2/trace/+ASM2_x000_28639240.trc
WARNING: cache read (retry) a corrupt block: group=4(XIFENFEI) 
 dsk=15 blk=258 disk=15 (XIFENFEI_0015) incarn=1717056824 au=113792 blk=2 count=1
Errors in file /u01/app/oracle/diag/asm/+asm/+ASM2/trace/+ASM2_x000_28639240.trc:
ORA-15196: invalid ASM block header [kfc.c:26368] [endian_kfbh] [2147483663] [258] [56 != 0]
ORA-15196: invalid ASM block header [kfc.c:26368] [endian_kfbh] [2147483663] [258] [56 != 0]
ERROR: cache failed to read group=4(XIFENFEI) dsk=15 blk=258 from disk(s): 15(XIFENFEI_0015)
ORA-15196: invalid ASM block header [kfc.c:26368] [endian_kfbh] [2147483663] [258] [56 != 0]
ORA-15196: invalid ASM block header [kfc.c:26368] [endian_kfbh] [2147483663] [258] [56 != 0]
NOTE: cache initiating offline of disk 15 group XIFENFEI
NOTE: process _x000_+asm2 (28639240) initiating offline of disk 
    15.1717056824 (XIFENFEI_0015) with mask 0x7e in group 4
NOTE: initiating PST update: grp = 4, dsk = 15/0x66583538, mask = 0x6a, op = clear
GMON updating disk modes for group 4 at 23 for pid 28, osid 28639240
ERROR: Disk 15 cannot be offlined, since diskgroup has external redundancy.
ERROR: too many offline disks in PST (grp 4)
Sun Nov 24 04:42:27 2019
NOTE: cache dismounting (not clean) group 4/0x0B08C40B (XIFENFEI) 
WARNING: Offline for disk XIFENFEI_0015 in mode 0x7f failed.
Sun Nov 24 04:42:27 2019
NOTE: halting all I/Os to diskgroup 4 (XIFENFEI)
NOTE: messaging CKPT to quiesce pins Unix process pid: 59441780, image: oracle@xifenfei2 (B000)
Sun Nov 24 04:42:27 2019
ERROR: ORA-15130 thrown in ARB0 for group number 4
Errors in file /u01/app/oracle/diag/asm/+asm/+ASM2/trace/+ASM2_arb0_50856926.trc:
ORA-15130: diskgroup "XIFENFEI" is being dismounted

至此两个节点的该磁盘组就陷入了不停的mount,然后dismount的轮流循环中.这里我们可以大概的分析出来,由于vg的磁盘组被写入了数据或者强制剔除的时候导致asm写入该文件的数据被破坏,导致后续的asm reblance遭遇坏块,然后直接dismount.对于该问题的解决方案,通过对对该磁盘组的acd和cod进行patch,让其不进行reblance,保持该磁盘组现在,稳定的mount状态,然后对其数据进行备份和重建该磁盘组.这个客户运气不错,vg中的asm disk磁盘写入较少,数据库运行正常.
对于这种情况,如果发生极端损坏,比如asm磁盘组无法mount,可以参考:找回ASM中数据文件
如果是asm的元数据大量损坏,无法通过asm字典级别恢复,可以通过参考:asm disk header 彻底损坏恢复

发表在 Oracle ASM | 标签为 , , , , | 评论关闭

KFED-00322: Invalid content encountered during block traversal: [kfbtTraverseBlock][Invalid OSM block type]

在oracle asm的使用过程中由于操作系统层面的错误操作导致asm disk 被破坏,这里列举了几种破坏之后的kfed报错现象(KFED-00322: Invalid content encountered during block traversal: [kfbtTraverseBlock][Invalid OSM block type])
asm mount 磁盘组报错(ORA-15040 ORA-15042)

SQL> alter diskgroup DATA mount;
alter diskgroup DATA mount
*
ERROR at line 1:
ORA-15032: not all alterations performed
ORA-15040: diskgroup is incomplete
ORA-15042: ASM disk "2" is missing from group number "2"

asm alert日志报错(ORA-15335 ORA-15066 ORA-15196等)

ORA-15335: ASM metadata corruption detected in disk group 'DATA'
ORA-15130: diskgroup "DATA" is being dismounted
ORA-15066: offlining disk "DATA_0002" in group "DATA" may result in a data loss
ORA-15196: invalid ASM block header [kfc.c:26368] [endian_kfbh] [2147483651] [48] [0 != 1]

kfed查看磁盘头报错
文件文件头(不光是disk header的4k,可能是连续的几个au,甚至更多)可能彻底损坏,一般kfed 读取都会看到KFED-00322: Invalid content encountered during block traversal: [kfbtTraverseBlock][Invalid OSM block type]之类错误

[oracle@fcomtaep2 disks]$ kfed read ASMRECO03
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
7FC18D899400 00000000 00000000 00000000 00000000 [................]
  Repeat 27 times
7FC18D8995C0 FEEE0001 0001FFFF FFFF0000 00000FFF [................]
7FC18D8995D0 00000000 00000000 00000000 00000000 [................]
  Repeat 1 times
7FC18D8995F0 00000000 00000000 00000000 AA550000 [..............U.]
7FC18D899600 20494645 54524150 00010000 0000005C [EFI PART....\...] <==== **** Here ******
7FC18D899610 BD82BBB3 00000000 00000001 00000000 [................]
7FC18D899620 0FFFFFFF 00000000 00000022 00000000 [........".......]
7FC18D899630 0FFFFFDE 00000000 FD8857E5 42D7B49B [.........W.....B]
7FC18D899640 0901FA87 6B3DB5AA 00000002 00000000 [......=k........]
7FC18D899650 00000080 00000080 FE48EB77 00000000 [........w.H.....]
7FC18D899660 00000000 00000000 00000000 00000000 [................]
  Repeat 25 times
7FC18D899800 EBD0A0A2 4433B9E5 B668C087 C79926B7 [......3D..h..&..]
7FC18D899810 5381F6DF 4626F988 0E4F468D D78D3B28 [...S..&F.FO.(;..]
7FC18D899820 000007A1 00000000 0FFFF85F 00000000 [........_.......]
7FC18D899830 00000000 00000000 00720070 006D0069 [........p.r.i.m.]
7FC18D899840 00720061 00000079 00000000 00000000 [a.r.y...........]
7FC18D899850 00000000 00000000 00000000 00000000 [................]
 Repeat 186 times
KFED-00322: Invalid content encountered during block traversal: [kfbtTraverseBlock][Invalid OSM block type][][0]

“EFI PART”是分区的元数据,一般是被分区导致asm disk损坏.

[ebernal@dbaasm new2]$ kfed read emcpowerl | head -25
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
2ABD671E9400 00000000 00000000 00000000 00000000 [................]
  Repeat 31 times
2ABD671E9600 4542414C 454E4F4C 00000001 00000000 [LABELONE........]
2ABD671E9610 E4E1DDB1 00000020 324D564C 31303020 [.... ...LVM2 001] <==== **** Here ******
2ABD671E9620 50365A77 71327874 34303156 4B4E6136 [wZ6Ptx2qV1046aNK]
2ABD671E9630 35395159 5147634C 487A5A38 63575A37 [YQ95LcGQ8ZzH7ZWc]
2ABD671E9640 00000000 00000019 00030000 00000000 [................]
2ABD671E9650 00000000 00000000 00000000 00000000 [................]
2ABD671E9660 00000000 00000000 00001000 00000000 [................]
2ABD671E9670 0002F000 00000000 00000000 00000000 [................]
2ABD671E9680 00000000 00000000 00000000 00000000 [................]
  Repeat 215 times
KFED-00322: Invalid content encountered during block traversal: [kfbtTraverseBlock][Invalid OSM block type][][0]

“LVM2 001” 是逻辑卷的名字,该asm disk很可能被做为lvm管理而被破坏

[ebernal@dbaasm tars]$ kfed read rhdisk16
kfbh.endian:                         65 ; 0x000: 0x41
kfbh.hard:                           73 ; 0x001: 0x49
kfbh.type:                           88 ; 0x002: *** Unknown Enum ***
kfbh.datfmt:                         32 ; 0x003: 0x20
kfbh.block.blk:              1111709260 ; 0x004: blk=1111709260
kfbh.block.obj:              1634861056 ; 0x008: file=131072
kfbh.check:                         119 ; 0x00c: 0x00000077
kfbh.fcn.base:                        0 ; 0x010: 0x00000000
kfbh.fcn.wrap:                        0 ; 0x014: 0x00000000
kfbh.spare1:                          0 ; 0x018: 0x00000000
kfbh.spare2:                          0 ; 0x01c: 0x00000000
2B6FE2AC1400 20584941 4243564C 61720000 00000077  [AIX LVCB..raw...] <==== **** Here ******
2B6FE2AC1410 00000000 00000000 00000000 00000000  [................]
2B6FE2AC1420 00000000 00000000 30300000 38306430  [..........000d08]
2B6FE2AC1430 30306131 34643030 30303030 31303030  [1a0000d400000001]
2B6FE2AC1440 61006533 766C6D73 7461645F 00003161  [3e.asmlv_data1..]
2B6FE2AC1450 00000000 00000000 00000000 00000000  [................]
        Repeat 2 times
2B6FE2AC1480 54000000 4D206575 20207961 31312037  [...Tue May  7 11]
2B6FE2AC1490 3A33343A 32203633 0A333130 00000000  [:43:36 2013.....]
2B6FE2AC14A0 65755400 79614D20 20372020 343A3131  [.Tue May  7 11:4]
2B6FE2AC14B0 34323A38 31303220 00000A33 44000000  [8:24 2013......D]
2B6FE2AC14C0 41313830 30303444 6D6D7900 02007900  [081AD400.ymm.y..]
2B6FE2AC14D0 0100E40C 656E6F4E 00000000 00000000  [....None........]
2B6FE2AC14E0 00000000 00000000 00000000 00000000  [................]
        Repeat 14 times
2B6FE2AC15D0 00000000 00000000 65310000 61653934  [..........1e49ea]
2B6FE2AC15E0 342E3862 00000000 00000000 00000000  [b8.4............]
2B6FE2AC15F0 00000000 00000000 00000000 00000000  [................]
  Repeat 224 times
KFED-00322: Invalid content encountered during block traversal: [kfbtTraverseBlock][Invalid OSM block type][][88]

这里的“AIX LVCB..raw” 是AIX OS volume 的元数据库,也就是说,asm disk 被作为了aix os层面破坏

[oracle@dbep2 disks]$ kfed read asm-disk3
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
06000000 00000000 00000000 00000000 00000000 [................]
  Repeat 25 times
0602100 51e2b7f6 00ed4e00 00000000 00000001  [...Q.N..........]
0602120 00000000 0000000b 00000100 0000003c [............<...]
0602140 00000242 0000007b 5d8468e7 6147782a [B...{....h.]*xGa]
0602160 d17851a2 327552e2 00000000 00000000 [.Qx..Ru2........]
0602200 00000000 00000000 3130752f 91a4f000 [......../u01....] <==== **** Here ******
0602220 ff8808e4 d5104cff 000000ac 00000100 [.....L..........]
0602240 00000000 00000000 00000000 09d18000 [................]
  Repeat 254 times
KFED-00322: Invalid content encountered during block traversal: [kfbtTraverseBlock][Invalid OSM block type][][88]

这里的/u01很可能表明该asm disk被文件系统覆盖

对于asm disk的各种破坏情况,如果是normal/high冗余,那么asm dg没有问题,可以考虑通过删除异常盘,然后重新加入;如果是外部冗余遭遇到asm disk 被破坏,一般asm disk 会dismount,而且无法正常mount,如果有备份的磁盘头,可以尝试还原磁盘头,mount 磁盘组,然后只读方式迁移数据;如果没有备份磁盘头或者还原之后也无法mount,可能需要通过一些额外的方式处理比如通过工具在asm dismount状态下恢复数据文件,甚至通过对asm block/oracle block碎片重组的方式恢复数据.参考相关文章:
oracle asm系列文章汇总
pvid=yes导致asm无法mount
asm disk header 彻底损坏恢复
分区无法识别导致asm diskgroup无法mount
oracle asm disk格式化恢复—格式化为ext4文件系统
oracle asm disk格式化恢复—格式化为ntfs文件系统
asm disk误设置pvid导致asm diskgroup无法mount恢复
分享oracleasm createdisk重新创建asm disk后数据0丢失恢复案例
ORA-15042: ASM disk “N” is missing from group number “M” 故障恢复
如果您遇到此类情况,无法解决请联系我们,提供专业ORACLE数据库恢复技术支持
Phone:13429648788    Q Q:107644445QQ咨询惜分飞    E-Mail:dba@xifenfei.com

发表在 Oracle ASM, Oracle备份恢复 | 标签为 , , , , , | 评论关闭