Problems with AHA2940UW and external RAID
Hubert Mantel
mantel at suse.de
Tue Feb 3 06:52:11 PST 1998
Hi,
On Tue, 3 Feb 1998, Doug Ledford wrote:
> > I disabled both Wide and Synchronous negotiation on the RAID drive (ID
> > 1) in the Adaptec's BIOS but this was not the solution. The same
> > problems occur.
> >
> > If I use DOS, I can communicate with the RAID-System, therefore this
> > seems to be a real problem of the Linux SCSI-driver and not of the
> > RAID drive or the controller. Also, I can use low-level format and
> > verify of the Adaptec's BIOS
>
> At this point we are down to only a few possibilities left. We've
> disabled sync and wide negotiation, so that isn't it, we aren't using
> tagged queueing yet so that isn't it, and the only things that should
> have taken place with that device on the bus are a TEST_UNIT_READY
> command and that's it. The command it hangs on is the INQUIRY command.
> We're kinda down to only a few things left. That device simply doesn't
> like some timing issue in the sequencer, it doesn't like our INQUIRY
> command from the mid level SCSI code, or possibly it has something to do
> with the Disconnect priviledge. The first two are the hardest ones to
> change, so let's try going into the Adaptec SCSI BIOS again and this time
> switch the disconnection enabled to disconnection disabled (or vice-versa
> depending on what it currently is) and see if that makes a difference.
FWIW, I had a very similar support question some time ago, and the problem
of the customer went away after upgrading the firmware of the RAID
controller.
As the original poster is from germany, I quote something from the
original answer:
----
Der Raid-Controller der Firma Infortrend Typ IFT-3102U muss um mit einem
Adaptec 2940 UW Controller zusammenarbeiten mindesten die
Firmwareversion 1.31 besitzen, sonst führt das SCSI-Target Inquiry zum
SCSI-BUS hang.
Das Update der Firmware ist via E-Mail unter support at infortrend.com zu
bestellen!
----
He also had the problem that the Inquiry led to a SCSI bus hang.
> Doug Ledford <dledford at dialnet.net>
hope it helps,
Hubert
More information about the aic7xxx
mailing list