Problems with AHA2940UW and external RAID

Andreas Welbers aw at ind.rwth-aachen.de
Tue Feb 3 04:26:03 PST 1998


Doug Ledford writes:
 > 
 > > (scsi0:0:1:0) Using wide (16 bit) transfers.
 > > (scsi0:0:1:0) Synchronous at 20.0MHz, offset 8.
 > > scsi : aborting command due to timeout : pid 3, scsi0, channel 0, id 1, 
 > >  lun 0 Inquiry 00 00 00 ff 00
 > 
 > OK...a few key notes here.  First, this is happening during the initial 
 > bus scan for devices.  During this scan, all commands are untagged 
 > commands and are carried out synchronously.  Second, it succeeds on the 
 > test unit ready command (the very first command sent) and then fails on 
 > the inquiry command.  It is possible that the device is having problems 
 > with negotiation in some way.  As a test to narrow down the source of the 
 > problem, you can go into the Adaptec BIOS (the CTRL-A message at bootup) 
 > and disable both Wide and Synchronous negotiation on the RAID drive (I 
 > know, it's useless to run such a fast device in that mode, but it helps 
 > to narrow down the problem).  If that doesn't help, then we really have a 
 > problem between this device and the standard inquiry command linux issues 
 > to all devices (most likely anyway).  When you've tried that, if it 
 > helps, then you can start enabling the sync and wide negotiation in the 
 > Adaptec BIOS again and slowly pump up the sync rate until it stops 
 > working again.  It may be that this device simply needs to be one step 
 > below Ultra-Wide speeds or something like that (minor timing variance 
 > between the card and this device causing a problem at full out Ultra-Wide 
 > speeds).  Try these things out and let me know what happens.
 >
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

 > 
Any other idea?








More information about the aic7xxx mailing list