Hi,
so, this weekend one of our clients cluster was killed due to a
catastrophic power failure. When power came back, the the admin-node
tried to recover a soft-RAID1 built from two "IBM-DTLA-307075 ATA DISK"
disk drives. Two hours into the rebuild we got a bunch of the following
messages:
hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hde: dma_intr: error=0x40 { UncorrectableError }, LBAsect=129011065,
sector=129010952
end_request: I/O error, dev 21:01 (hde), sector 129010952
As a consequence, the drive with the errors was disabled by the RAID
software. Now we are running with only one drive.
The mesaages themselves seem to indicate problems on the hde drive,
which could be easily verified by doing a partial badblocks scan on the
region of sectors/blocks in question. As a result I now have a list of
bad blocks for that disk.
The question now is: how do I proceed? First of all, a backup. That is
the no-brainer. But how do I rebuild the RAID1 and the ReiserFS on it
without using the bad blocks?
Thanks
Martin
--
------------------------------------------------------------------
Martin Knoblauch | email: [EMAIL PROTECTED]
TeraPort GmbH | Phone: +49-89-510857-309
C+ITS | Fax: +49-89-510857-111
http://www.teraport.de | Mobile: +49-170-4904759
