Quoting Stephen Lord <lord@xxxxxxx>:
> Unfortunately, some hardware is just designed to rely on write
> caching. It does seem a little odd that you lost the superblock
> though. This is put there by mkfs, and is always present, so how
> powering down the device is corrupting it I do not know. That
> seems like an issue with the 3ware firmware to me.
Yes, as with _all_ RAID controllers, you _must_ make sure the driver and
firmware "match up." _You_ must be _vigilant_ to do this as OS upgrades and
kernel updates in newer versions can introduce newer drivers than require newer
firmware.
CASE-IN-POINT: _Always_ investigate the driver version in a newer kernel or OS,
and make sure you have a compatible firmware, _before_ upgrading it.
My Linux /var filesystem (XFS) was heavily corrupted (~50% file loss) after I
upgraded to Windows XP and loaded a newer 3Ware ATA RAID driver than the
firmware was designed for. Yes, the driver in another OS messed with blocks
outside its partitions, because the firmware was too old.
Had a /home filesystem (Ext3) filesystem was moderately corrupted (~100 files) I
upgraded an ICP-Vortex SCSI RAID kernel driver without upgrading the firmware,
and failed disk detection took 24 seconds before it realized it needed to stop
writing to it.
--
Bryan J. Smith, E.I. Contact Info: http://thebs.org
A+/i-Net+/Linux+/Network+/Server+ CCNA CIWA CNA SCSA/SCWSE/SCNA
---------------------------------------------------------------
'In the new commercials for MSN 8, a guy dressed in a Microsoft
Butterfly costume drops out of a "cocoon" ... nobody at
Microsoft ... knows that a butterfly doesn't emerge from a
cocoon, but a chrysalis -- moths come from cocoons ... "this
should not be all that surprising since Microsoft seems to
have problems recognizing bugs ..."' -- Cringely
|