I send two disks back to my customer, because there were damaged and got
timeouts all the day under Linux.
Our customer called me back same day and ask me what is wrong with
disks? He had tested them under NT again and they were running very
fine...
Greetings, Dietmar
[EMAIL PROTECTED] wrote:
>
> Charles Galpin said ...
> >
> > yes, that's where I posted this. I didn't notice that this is now going to
> > and linux-raid. How did that happen?
> >
> > I believe the termination is fine. I have contacted both adaptec and IBM,
> > and they both say that NT is very pick y about cabling and termination, and
> > if that works then they think its OK. They of course suspect the linux
> > driver. I am beginning to belive this.
>
> Boy, that is the truth. NT is *very* tight on the timing and anything
> marginal (ie: anything that can cause errors, retries or delays) will
> cause problems. I used to have an old AHA1542 controller that NT just
> would not use reliably, yet Linux never had any problems with it.
>
> You might also try removing the external terminator and enabling the
> drives internal termination (with power provided by the drive).
>
> > Still desperate to resolve this. I tried attaching an IDE drive so I could
> > get a base install and upgrade it to a 2.2 kernel and the latest scsi
> > drivers since I have corresponeded with someone who has the 2940U2W working
> > fine, but added this card to an existing system running 2.2.6.
>
> Let me ask: does the Ultra bus work without any problems? Is it the
> entire card, or just the LVD bus thats having problems? I've got an
> 2940U2W, with several HDs, Tape drives and CDROMs on it (but only on the
> Ultra, not the LVD bus). Only had one problem with it and that was one
> of the middle drives had termination enabled by mistake. Now, it's sold
> as a rock. But, I'm using aic7xxx-5.1.13 and a 2.0.36 kernel.
>
> If the Ultra bus works without any problems, then perhaps the IBM drive
> is defective (it does happen).
>
> Timeouts typically indicate a loss of signal. I know you've checked the
> termination, but check it again and also check the parity setting for the
> drive and card. If throttling down to 5.0Mb doesn't work, then try the
> SCSI BIOS's Verify utility. See if it can verify the drive. If not,
> then you've definitely got a cable or termination problem. If it can,
> especially at the higher speeds, then it's a software issue.
>
> > Anyway, when I add the drive, enable IDE support in the bios, it does not
> > get detected by my MOBO, and then the system won't boot off the scsi drive
> > anymore. I tried as master on both primary and secondary channel - no
> > difference.
>
> Hmm! This might mean theirs a lack of free IRQs to allocate to the SCSI
> card, or that the Motherboard itself has a problem. Make sure that you
> have plenty of non ISA allocated IRQs available for the BIOS to assign.
>
> I do recall once a friend was having a horrendous problem with a video
> controller. We ended up disassembling the entire computer and found that
> the Motherboard wasn't mounted correctly. One of the ground screws was
> forced into place and cause stress on the Motherboard. Removing that
> screw and reseating the board fixed the problem.
>
> > Other than somethign liek the above, I am out of ideas to get this working.
> > Anyone know of a distro out there with boot disks using a 2.2 kernel?
>
> Good luck!
>
> > tia
> > charles
>
> --
> Peter A. Castro ([EMAIL PROTECTED]) or ([EMAIL PROTECTED])
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
> the body of a message to [EMAIL PROTECTED]
--
"Don't fear the Gates" (a little like Blue Oyster Cult)
Dietmar Stein, Systemadministrator UNIX/Linux
[EMAIL PROTECTED], [EMAIL PROTECTED]
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to [EMAIL PROTECTED]
