[Linux-scsi] HELP : ADAPTEC 2940 UW Pro an ongoing saga

[Jeff Hill]

Well, I can't offer any direct help but I can empathize. It took me four
months to get a 2940U2W working with aic7xxx. I spent 140 hours working
on it, begging Doug Ledford (the aic7xxx maintainer) for assistance. He
had offered (on the list) to help those with problems similar to mine,
which gave me encouragement, and then he never responded to my queries
(pure torture). I offered to pay him and received no response. While I
didn't end up crying, I did beat my head against the wall and cursed
profusely.

That said, you mention a SELTO problem. There is a boot option to change
the seltime that might help (see the README.aic7xxx file). As well, try
the various other boot command line options. 

Below is a short list of some of the things I tried getting the aic7xxx
to work properly. 

While this may not help, perhaps it will give you some hope.

Regards,

Jeff Hill
-----------

 * upgraded the AIC7xxx driver from 5.1.21 to 5.1.30, testing 
   each version;

 * tried kernels from 2.2.10 to 2.2.16;

 * added MTRR (Memory Type Range Register) support in kernel;

 * replaced ATI Mach64 card with cheap no-name to test 
   video card conflict;

 * replaced OEM SCSI cable with custom teflon LVD cable with 
   active negation terminator (cost $125US);

 * replaced ethernet card;

 * removed the IDE drives and kernel support for IDE;

 * swapped positions of cards;

 * lowered the front bus speed from 100.3Mhz to 88.3 
   (underclocking CPU);

 * checked IRQs for conflicts; 

 * compiled the kernel with and without tagged command queuing, 
   with more and fewer max commands;

 * disabled virtually everything possible in the motherboard bus 
   to avoid conflicts;

 * removed my RAID-1 and added it back in (unable to fully 
   test this);

 * did RAM tests on my ECC chips;

 * added verbose debugging boot option;

 * repeatedly searched the archives of mailing lists for aic7xxx,  
   linux-scsi, Adaptec, and ASUS -- no luck.   

 * added disk drive fans for extra cooling;

 * tested power supply to ensure clean voltage;

 * reduced SCSI controller speed on drives to 40MB from 80MB;

 * set up remote logging to catch any error messages missed when
   system freezes;



 
> Michael Parkinson [michael at intellnet.net.uk] wrote:
> > Dear All,
> >
> > Starting off with a 29160 and 2 IBM 160 drives I have now been reduced to a
> > 2940UW Pro with the same hard drives.
> >
> > I have patched the 2.0.36 kernel to no avail, I have upgraded to the 2.2.16
> > and further the 2.2.17 kernel.
> >
> > I have patched all the patches on the aic7xxx site.
> >
> > I am near tears.
> >
> > I cant get the system to boot now on the 2.2.17 as It bombs out with a
> > SELTO problem, 2.2.16 works sometimes but then falls over with the scsi bus
> > reseting.  The 2.0.36 patched kernel works fine except when I do something
> > active on the harddrive like compliling a kernel at which point the file
> > system slowly starts corrupting.
> >
> > PLEASE PLEASE PLEASE HAS ANYONE MANAGED TO GET A WORKING STABLE SYSTEM WITH
> > ANY OF THE FOLLOWING SCSI CARDS AND WHAT DO I NEED TO DO TO ACHIEVE IT:
> >
> > ADAPTEC 29160
> > ADAPTEC 2940 U2W
> > ADAPTEC 2940UW Pro
> > ADAPTEC 2940UW
> >
> > I have reached the point of begging humbly.
> >
> > Will supply any info required.
> >
> > With thanks
> >
> > Michael
> >
> >
> > ====================================================
> > http://www.ishop.co.uk/
> > Build a shop on-line.
> > Buy online.
> > ====================================================
> >
> > Michael Parkinson BSc.(Hons)
> > Project Manager
> > Intellnet Limited
> > Heath House
> > 51 Dane Street
> > Bishops Stortford.
> > Hertfordshire
> > CM23 3BT
> >

------------------------------------------------------------
------  HR On-Line:  The Network for Workplace Issues ------
http://www.hronline.com - Ph:416-604-7251 - Fax:416-604-4708
------------------------------------------------------------


To Unsubscribe: send mail to majordomo at FreeBSD.org
with "unsubscribe aic7xxx" in the body of the message