SW Raid 1 se rozpada

[Zkoumalek]

Dobry den.

Mam na serveru Mandrake Linux 10.0, pred rokem jsem na nem rozbehl sw raid 1 na 2 ide discich Maxtor 120GB. Pouzil jsem jako dalsi ide radic kartu s cipem SIL0680ACL144. Bohuzel se asi po mesici raid rozpadl. Znovu jsem odebral a pridal zlobivy disk do raidu a vse bezelo dalsi pul rok OK. Nyni se zase raid rozpadl. Chyb je plny log, jeste ke vsemu se pocochnil i ext3, ktery je na raidu nainstalovany, takze jsem musel nechat souborovy system opravit. Badblocks se tvari, ze je vse v poradku. Nakonec jsem vyhodil pridavnou IDE kartu a pripojil jsem to primo na radic na desce. Nevi nekdo, proc se raid stale rozpada? Zalezi na souborovem systemu, ktery na raidu mam? Ma nekdo zkusenosti s timto SIL cipem a IDE radicem? Jadro mam puvodni  2.4.25-2mdk, nemuze byt problem v jadru?

Predem dekuji za pomoc.

T. Zkoumalek

Cast logu:
smartd[1333]: Device: /dev/hde, 63 Offline uncorrectable sectors
smartd[1333]: Device: /dev/hdg, 750 Currently unreadable (pending) sectors
smartd[1333]: Device: /dev/hdg, 750 Offline uncorrectable sectors
kernel: hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
kernel: hde: dma_intr: error=0x40 { UncorrectableError }, LBAsect=136843775, sector=136843712
kernel: end_request: I/O error, cmd 0 dev 21:01 (hde), sector 136843712
kernel: raid1: ide/host2/bus0/target0/lun0/part1: rescheduling block 136843712
kernel: raid1: ide/host2/bus0/target0/lun0/part1: unrecoverable I/O read error for block 136843712
kernel: EXT3-fs error (device md(9,0)): ext3_readdir: directory #8552475 contains a hole at offset 0
kernel: hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
kernel: hde: dma_intr: error=0x40 { UncorrectableError }, LBAsect=214175847, sector=214175784
kernel: end_request: I/O error, cmd 0 dev 21:01 (hde), sector 214175784
kernel: raid1: ide/host2/bus0/target0/lun0/part1: rescheduling block 214175784
kernel: raid1: ide/host2/bus0/target0/lun0/part1: unrecoverable I/O read error for block 214175784
kernel: EXT3-fs error (device md(9,0)): ext3_readdir: directory #13385734 contains a hole at offset 0
kernel: hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
kernel: hde: dma_intr: error=0x40 { UncorrectableError }, LBAsect=145494383, sector=145494320
kernel: end_request: I/O error, cmd 0 dev 21:01 (hde), sector 145494320
kernel: raid1: ide/host2/bus0/target0/lun0/part1: rescheduling block 145494320
kernel: raid1: ide/host2/bus0/target0/lun0/part1: unrecoverable I/O read error for block 145494320
kernel: EXT3-fs error (device md(9,0)): ext3_readdir: directory #9093134 contains a hole at offset 0
kernel: hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
kernel: hde: dma_intr: error=0x40 { UncorrectableError }, LBAsect=230430427, sector=230430272

____________________________________________________________
http://www.seznam.cz - e-mailová schránka 2000 MB