Dotaz: Read-write problem s LVM (3x250GB)

[fixinko]

Zdravim, mam na jadre 2.6.19.2 spravene LVM na 3x250GB diskoch, ktore su pripojene na IDE PCI radic ktory sa v linuxe hlasi ako IT8212. Pri boote sa to hlasi vsetko takto:

[    8.241886] IT8212: IDE controller at PCI slot 0000:00:10.0
[    8.308700] IT8212: chipset revision 19
[    8.354667] it821x: controller in smart mode.
[    8.406858] IT8212: 100% native mode on irq 17
[    8.460115]     ide1: BM-DMA at 0xbca0-0xbca7, BIOS settings: hdc:DMA, hdd:pio
[    8.544767]     ide2: BM-DMA at 0xbca8-0xbcaf, BIOS settings: hde:pio, hdf:pio
[    8.809492] hdc: Hitachi HDT725025VLAT80, ATA DISK drive
[    9.179577] hdc: Performing identify fixups.
[    9.230708] ide1 at 0xbc88-0xbc8f,0xbc9a on irq 17
[    9.553233] hde: Hitachi HDT725025VLAT80, ATA DISK drive
[    9.923343] hde: Performing identify fixups.
[    9.974534] ide2 at 0xbc90-0xbc97,0xbc9e on irq 17
[   10.032503] IT8212: IDE controller at PCI slot 0000:00:12.0
[   10.099306] IT8212: chipset revision 19
[   10.145229] it821x: controller in smart mode.
[   10.145246] IT8212: 100% native mode on irq 16
[   10.145275]     ide3: BM-DMA at 0xbc70-0xbc77, BIOS settings: hdg:pio, hdh:pio
[   10.145342]     ide4: BM-DMA at 0xbc78-0xbc7f, BIOS settings: hdi:pio, hdj:pio
[   10.409999] hdg: Hitachi HDT725025VLAT80, ATA DISK drive
[   10.717035] hdg: Performing identify fixups.
[   10.768193] ide3 at 0xbc58-0xbc5f,0xbc6a on irq 16
[   11.866693] hda: max request size: 128KiB
[   11.914672] hda: 8448300 sectors (4325 MB), CHS=8940/15/63, UDMA(33)
[   11.991490]  hda: hda1
[   12.172540] hdc: max request size: 128KiB
[   12.220570] hdc: 488397168 sectors (250059 MB) w/7372KiB Cache, CHS=30401/255/63, BUG DMA OFF
[   12.323657]  hdc:hdc: recal_intr: status=0x51 { DriveReady SeekComplete Error }
[   12.412164] hdc: recal_intr: error=0x04 { DriveStatusError }
[   12.412188] ide: failed opcode was: unknown
[   23.476291] hdc: irq timeout: status=0xd0 { Busy }
[   23.533806] ide: failed opcode was: unknown
[   23.633224] ide1: reset: master: ECC circuitry error
[   24.721720]  unknown partition table
[   24.765212] hde: max request size: 128KiB
[   24.813254] hde: 488397168 sectors (250059 MB) w/7372KiB Cache, CHS=30401/255/63, BUG DMA OFF
[   24.916287]  hde:hde: recal_intr: status=0x51 { DriveReady SeekComplete Error }
[   25.004640] hde: recal_intr: error=0x04 { DriveStatusError }
[   25.004661] ide: failed opcode was: unknown
[   36.069895] hde: irq timeout: status=0xd0 { Busy }
[   36.127426] ide: failed opcode was: unknown
[   36.227834] ide2: reset: master: ECC circuitry error
[   37.318164]  unknown partition table
[   37.361681] hdg: max request size: 128KiB
[   37.409672] hdg: 488397168 sectors (250059 MB) w/7372KiB Cache, CHS=30401/255/63, BUG DMA OFF
[   37.512611]  hdg:hdg: recal_intr: status=0x51 { DriveReady SeekComplete Error }
[   37.601167] hdg: recal_intr: error=0x04 { DriveStatusError }
[   37.669164] ide: failed opcode was: unknown
[   48.665500] hdg: irq timeout: status=0xd0 { Busy }
[   48.723023] ide: failed opcode was: unknown
[   48.821435] ide3: reset: master: ECC circuitry error
[   49.908581]  unknown partition table

Najskor vsetko fungovalo normalne, ale teraz je ten problem, ze ked chcem na to pole nieco zapisat tak vidim v dmesg taketo nieco:

[  661.978174] end_request: I/O error, dev hdc, sector 345001432
[  661.978195] __journal_remove_journal_head: freeing b_frozen_data
[  661.978218] __journal_remove_journal_head: freeing b_committed_data
[  661.978250] __journal_remove_journal_head: freeing b_frozen_data
[  661.978218] __journal_remove_journal_head: freeing b_committed_data
[  661.978250] __journal_remove_journal_head: freeing b_frozen_data
[  661.978272] end_request: I/O error, dev hdc, sector 345004928
[  661.978301] __journal_remove_journal_head: freeing b_frozen_data
[  661.978322] end_request: I/O error, dev hdc, sector 345005456
[  661.978352] __journal_remove_journal_head: freeing b_frozen_data
[  661.978373] end_request: I/O error, dev hdc, sector 345005480
[  661.978402] __journal_remove_journal_head: freeing b_frozen_data
[  661.978419] __journal_remove_journal_head: freeing b_committed_data
[  661.978528] end_request: I/O error, dev hdc, sector 345009576
[  661.978574] end_request: I/O error, dev hdc, sector 345011088
[  685.022347] end_request: I/O error, dev hdc, sector 344932736
[  685.022456] EXT3-fs error (device dm-0): ext3_readdir: directory #21544961 contains a hole at offset 0
[  687.822486] end_request: I/O error, dev hdc, sector 344932736
[  687.822591] EXT3-fs error (device dm-0): ext3_readdir: directory #21544961 contains a hole at offset 0
[  688.475297] end_request: I/O error, dev hdc, sector 344932736
[  688.475411] EXT3-fs error (device dm-0): ext3_readdir: directory #21544961 contains a hole at offset 0
[  694.252807] end_request: I/O error, dev hdc, sector 344932736
[  694.252917] EXT3-fs error (device dm-0): ext3_readdir: directory #21544961 contains a hole at offset 0
[  704.578133] end_request: I/O error, dev hdc, sector 344932736
[  704.581904] EXT3-fs error (device dm-0): ext3_readdir: directory #21544961 contains a hole at offset 0

a filesystem sa remountne z read-write na read-only. Ked dam reboot tak v dmesg vidim taketo nieco:

Unmounting local file systems.
[  1244.285765] __journal_remove_journal_head: freeing b_frozen_data
[  1244.357777] __journal_remove_journal_head: freeing b_frozen_data
[  1244.429660] __journal_remove_journal_head: freeing b_committed_data
[  1244.504734] __journal_remove_journal_head: freeing b_frozen_data
[  1244.576644] __journal_remove_journal_head: freeing b_frozen_data
[  1244.648562] __journal_remove_journal_head: freeing b_committed_data
[  1244.723872] end_request: I/O error, dev hdc, sector 4896
[  1244.787410] printk: 12 messages suppressed.
[  1244.837502] Buffer I/O error on device dm-0, logical block 564
[  1244.907351] lost page write due to I/O error on dm-0
[  1244.992297] end_request: I/O error, dev hdc, sector 0
  /dev/hdc: read failed after 0 of 4096 at 0: Input/output error
[  1245.178891] end_request: I/O error, dev hdc, sector 0
[  1245.239374] Buffer I/O error on device hdc, logical block 0                                                                                                [  1245.306191] end_request: I/O error, dev hdc, sector 8                                                                                                      [  1245.366595] Buffer I/O error on device hdc, logical block 1
[  1245.433377] end_request: I/O error, dev hdc, sector 16
[  1245.494866] Buffer I/O error on device hdc, logical block 2
[  1245.561636] end_request: I/O error, dev hdc, sector 24
[  1245.623136] Buffer I/O error on device hdc, logical block 3
[  1245.690074] end_request: I/O error, dev hdc, sector 0
 [  1245.690111] Buffer I/O error on device hdc, logical block 0
/dev/hdc: read f[  1245.819623] end_request: I/O error, dev hdc, sector 488396928
ailed after 0 of[  1245.905054] end_request: I/O error, dev hdc, sector 0
 4096 at 0: Input/output error
  /dev/hdc: read failed after 0 [  1246.036117] end_request: I/O error, dev hdc, sector 384
of 4096 at 250059227136: Input/output error
  /[  1246.157549] end_request: I/O error, dev hdc, sector 0
dev/hdc: read fa[  1246.225759] end_request: I/O error, dev hdc, sector 384
iled after 0 of [  1246.310322] end_request: I/O error, dev hdc, sector 0
4096 at 0: Input[  1246.382055] end_request: I/O error, dev hdc, sector 384
/output error
 [  1246.463699] end_request: I/O error, dev hdc, sector 0
 /dev/fileserver[  1246.538478] end_request: I/O error, dev hdc, sector 384
/data: read fail[  1246.616124] end_request: I/O error, dev hdc, sector 0
ed after 0 of 40[  1246.694681] end_request: I/O error, dev hdc, sector 384
96 at 0: Input/output error
  Couldn't find device with uuid '29TD1l-70ab-aSe8-KJ34-m2Cp-8Rwp-33suzk'.
  Couldn't find all physical volumes for volume group fileserver.
  /dev/hdc: read failed after 0 of 4096 at 0: Input/output error
  /dev/fileserver/data: read failed after 0 of 4096 at 0: Input/output error
  Couldn't find device with uuid '29TD1l-70ab-aSe8-KJ34-m2Cp-8Rwp-33suzk'.
  Couldn't find all physical volumes for volume group fileserver.
  /dev/hdc: read failed after 0 of 4096 at 0: Input/output error
  /dev/fileserver/data: read failed after 0 of 4096 at 0: Input/output error
  /dev/hdc: read failed after 0 of 4096 at 0: Input/output error
  /dev/fileserver/data: read failed after 0 of 4096 at 0: Input/output error
  Couldn't find device with uuid '29TD1l-70ab-aSe8-KJ34-m2Cp-8Rwp-33suzk'.
  Couldn't find all physical volumes for volume group fileserver.
  /dev/hdc: read failed after 0 of 4096 at 0: Input/output error

SMART nehlasi ziadne chyby, fsck prebehne s takymto vysledkom:

root@(none)[tts/0]:~>fsck.ext3 -c -y -f -v /dev/fileserver/data
e2fsck 1.38 (30-Jun-2005)
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
    53497 inodes used (0%)
     2141 non-contiguous inodes (4.0%)
          # of inodes with ind/dind/tind blocks: 16594/6038/5
 74350352 blocks used (40%)
        0 bad blocks
       10 large files
    50247 regular files
     3240 directories
        0 character device files
        0 block device files
        0 fifos
        0 links
        1 symbolic link (1 fast symbolic link)
        0 sockets
    53488 files

Skusal som vymenit aj IDE PCI radic, aj kable, ale bez uspechu. Kde moze byt problem? Moze byt problem v disku? Vdaka.

Odpovědi

[Pindal]

end_request: I/O error, dev hdc, sector 345009576

Problem je v disku. Opravdu v SMART logu nic neni? Zkuste take SMART testy.

[fixinko]

SMART nespomina ziadne errory... Ked som tie disky daval do prevadzky (cca na konci januara 2007) tak som pustil na to mkfs.ext3 -c -c a to nehlasilo ziadne chyby... Nemoze to byt problem zleho vytvorenia LVM? Postupoval som podla http://tldp.org/HOWTO/LVM-HOWTO/.