Erik Jakobsen
Humm, I am suspecting a drive problem or might it be the DMA SuSE script. Are you enabling that script? I do not use 7.3, but I do use this on SLES, which is based on 7.2.
Ok. No I do not enable that script. Not AFAIK. How can I see that ?.
DMA is enabled automatically by the default SuSE kernel in 7.3. The "backup" kernel (vmlinuz.suse) doesn't use DMA so you may try this one instead. It can be selected by LILO and checked by hdparm: # /sbin/hdparm -v /dev/hda /dev/hda: multcount = 16 (on) I/O support = 0 (default 16-bit) unmaskirq = 0 (off) using_dma = 1 (on) keepsettings = 0 (off) nowerr = 0 (off) readonly = 0 (off) readahead = 8 (on) geometry = 2491/255/63, sectors = 40031712, start = 0 The variable START_IDEDMA="yes" is in /etc/rc.config in SuSE 7.2 but not in 7.3, so may forget about it.
If not, I would suspect the drive. I have one right now that is not dead, but takes forever seeking over some sectors, you may have a similar problem.
I have the same problem. Its a new Seagate 40GB.
IBM provides a utility called Drive Fitness Test (DFT) which checks its disks for HW errors. It's quite good if you want to make sure the disk is really bad. I wonder if there is a similar utility for Seagate disks. (I haven't searched for it.) -- Alexandr.Malusek@imv.liu.se
On Friday 28 December 2001 11:08, Alexandr Malusek wrote: Hi Alexandr.
DMA is enabled automatically by the default SuSE kernel in 7.3. The "backup" kernel (vmlinuz.suse) doesn't use DMA so you may try this one instead. It can be selected by LILO and checked by hdparm:
I think I have the option to select vmlinuz.suse at boot_up ?.
# /sbin/hdparm -v /dev/hda
/dev/hda: multcount = 16 (on) I/O support = 0 (default 16-bit) unmaskirq = 0 (off) using_dma = 1 (on) keepsettings = 0 (off) nowerr = 0 (off) readonly = 0 (off) readahead = 8 (on) geometry = 2491/255/63, sectors = 40031712, start = 0
Here's mine: # /sbin/hdparm -v /dev/hda /dev/hda: multcount = 16 (on) I/O support = 1 (32-bit) unmaskirq = 1 (on) using_dma = 1 (on) keepsettings = 0 (off) nowerr = 0 (off) readonly = 0 (off) readahead = 8 (on) geometry = 4865/255/63, sectors = 78165360, start = 0
The variable START_IDEDMA="yes" is in /etc/rc.config in SuSE 7.2 but not in 7.3, so may forget about it.
Yes nothing to do there.
IBM provides a utility called Drive Fitness Test (DFT) which checks its disks for HW errors. It's quite good if you want to make sure the disk is really bad. I wonder if there is a similar utility for Seagate disks. (I haven't searched for it.)
Ok. I'll have a look for that tool on the Seagate homepage. -- Med venlig hilsen / Best regards /vy73 de oz4kk. Erik Jakobsen - erik@urbakken.dk SuSE Linux 7.3.
I'll speculate that we're seeing these errors on so many machines
about the time SuSE (and other *nixes) started supporting DMA. It could
simply be crappy hardware, but nobody noticed as long as DMA
wasn't enabled. Sort of like the way MS-DOS will run fine on crap
hardware that will crash immediately if you install Linux or Windows
on it.
I'm also wondering: has anyone seen these errors on scsi disks, or is
it strictly an IDE problem?
- Robert Storey
On Fri, 28 Dec 2001 11:57:46 +0100
Erik Jakobsen
On Friday 28 December 2001 11:08, Alexandr Malusek wrote:
Hi Alexandr.
DMA is enabled automatically by the default SuSE kernel in 7.3. The "backup" kernel (vmlinuz.suse) doesn't use DMA so you may try this one instead. It can be selected by LILO and checked by hdparm:
I think I have the option to select vmlinuz.suse at boot_up ?.
# /sbin/hdparm -v /dev/hda
/dev/hda: multcount = 16 (on) I/O support = 0 (default 16-bit) unmaskirq = 0 (off) using_dma = 1 (on) keepsettings = 0 (off) nowerr = 0 (off) readonly = 0 (off) readahead = 8 (on) geometry = 2491/255/63, sectors = 40031712, start = 0
Here's mine:
# /sbin/hdparm -v /dev/hda
/dev/hda: multcount = 16 (on) I/O support = 1 (32-bit) unmaskirq = 1 (on) using_dma = 1 (on) keepsettings = 0 (off) nowerr = 0 (off) readonly = 0 (off) readahead = 8 (on) geometry = 4865/255/63, sectors = 78165360, start = 0
The variable START_IDEDMA="yes" is in /etc/rc.config in SuSE 7.2 but not in 7.3, so may forget about it.
Yes nothing to do there.
IBM provides a utility called Drive Fitness Test (DFT) which checks its disks for HW errors. It's quite good if you want to make sure the disk is really bad. I wonder if there is a similar utility for Seagate disks. (I haven't searched for it.)
Ok. I'll have a look for that tool on the Seagate homepage.
-- Med venlig hilsen / Best regards /vy73 de oz4kk. Erik Jakobsen - erik@urbakken.dk SuSE Linux 7.3.
-- To unsubscribe send e-mail to suse-linux-e-unsubscribe@suse.com For additional commands send e-mail to suse-linux-e-help@suse.com Also check the FAQ at http://www.suse.com/support/faq and the archives at http://lists.suse.com
Nope, nothing with SCSI AFAIK, and that is part of the reason I have not
been able to offer more clues, I do not use IDE much, the IBM Deskstar 40gig
I bought to hold divx files is sitting here on my desk dead. SCSI costs
more, but the headaches of IDE are not worth it. Sanity still makes SCSI
the defacto on servers.
Regards,
Jon
----- Original Message -----
From: "Robert Storey"
I'll speculate that we're seeing these errors on so many machines about the time SuSE (and other *nixes) started supporting DMA. It could simply be crappy hardware, but nobody noticed as long as DMA wasn't enabled. Sort of like the way MS-DOS will run fine on crap hardware that will crash immediately if you install Linux or Windows on it.
I'm also wondering: has anyone seen these errors on scsi disks, or is it strictly an IDE problem?
- Robert Storey
On Fri, 28 Dec 2001 11:57:46 +0100 Erik Jakobsen
wrote: On Friday 28 December 2001 11:08, Alexandr Malusek wrote:
Hi Alexandr.
DMA is enabled automatically by the default SuSE kernel in 7.3. The "backup" kernel (vmlinuz.suse) doesn't use DMA so you may try this one instead. It can be selected by LILO and checked by hdparm:
I think I have the option to select vmlinuz.suse at boot_up ?.
# /sbin/hdparm -v /dev/hda
/dev/hda: multcount = 16 (on) I/O support = 0 (default 16-bit) unmaskirq = 0 (off) using_dma = 1 (on) keepsettings = 0 (off) nowerr = 0 (off) readonly = 0 (off) readahead = 8 (on) geometry = 2491/255/63, sectors = 40031712, start = 0
Here's mine:
# /sbin/hdparm -v /dev/hda
/dev/hda: multcount = 16 (on) I/O support = 1 (32-bit) unmaskirq = 1 (on) using_dma = 1 (on) keepsettings = 0 (off) nowerr = 0 (off) readonly = 0 (off) readahead = 8 (on) geometry = 4865/255/63, sectors = 78165360, start = 0
The variable START_IDEDMA="yes" is in /etc/rc.config in SuSE 7.2 but not in 7.3, so may forget about it.
Yes nothing to do there.
IBM provides a utility called Drive Fitness Test (DFT) which checks its disks for HW errors. It's quite good if you want to make sure the disk is really bad. I wonder if there is a similar utility for Seagate disks. (I haven't searched for it.)
Ok. I'll have a look for that tool on the Seagate homepage.
-- Med venlig hilsen / Best regards /vy73 de oz4kk. Erik Jakobsen - erik@urbakken.dk SuSE Linux 7.3.
-- To unsubscribe send e-mail to suse-linux-e-unsubscribe@suse.com For additional commands send e-mail to suse-linux-e-help@suse.com Also check the FAQ at http://www.suse.com/support/faq and the archives at http://lists.suse.com
-- To unsubscribe send e-mail to suse-linux-e-unsubscribe@suse.com For additional commands send e-mail to suse-linux-e-help@suse.com Also check the FAQ at http://www.suse.com/support/faq and the archives at http://lists.suse.com
participants (4)
-
Alexandr Malusek
-
Erik Jakobsen
-
Jon
-
Robert Storey