Re: IDE disk corruption - A7M266-D; which kernel patches? or other solution
Pigeon said:
> I think there are a lot of VIA chipsets out there but most people
> don't have problems; presumably some other factor is needed to make the
> bug show up. Wonder what?
not entirely related but thought to mention ..
about a year and a half ago I was looking for a MB to get an Athlon 1300
to run with. I wanted a MB that supported SDRAM, and at least at the time
there were very few. I settled originally on the Asus A7A266. I transplanted
everything from my P3-800 except the MB/CPU/RAM to the new A7A266 including
the PCI Promise ATA/100 controller card. I purchased 3x256MB Kingston PC133
memory sticks for the board.
But the system suffered from massive disk curroption. I traced it down to
the point where I could reproduce it with a few commands:
mke2fs /dev/hde5
mount /dev/hde5 /mnt
umount /mnt
e2fsck /dev/hde5
(where /dev/hde5 was whatever partition I was testing and /mnt was the
mountpoint, I don't remember exactly what directory I was using but it
doesn't matter).
running the e2fsck after the umount showed MASSIVE filesystem curroption,
and I hadn't even written any files to the disk! just mounted it and
immediately unmounted it.
eventually I gave up and replaced the board with a Tyan board(which maxxed
out at 1300, so I can't upgrade the CPU). That was about a year and a half
ago, not a glitch since. Well excluding having my IBM IDE drives fail, which
I later replaced with a 9GB SCSI disk. So sick of having IDE drives fail!
one of my former co workers later told me(~6 months ago) that the particular
board I have(A7A266) apparently had some problems, I'm not sure what rev
was affected and if I had the affected rev or not. I never did try the onboard
IDE controller.
crazy shit.
nate
Reply to: