:)
>manual termination ENABLED on the AHA-2940 Ultra, and it works, many
>thanks! It's up and running linux 2.0.33-SMP for >3 days, that's good
>news. Btw, if I DISABLED termination and attached a (physical) terminator
>instead, is that also ok?
Excellent :) As for the external terminator, I wouldn't worry about it if
the card is terminating things correctly now. Just make sure you have both
ends of the SCSI bus terminated (the card on one end and the last device on
the other end of the cable) and you should be fine. If you have two cables
hooked up to that card, then the termination on the card should be turned
off, but given the problems you cited, I would guess that the devices at the
ends of the two cables aren't properly terminating things if you do have two
cables hooked up and it didn't work right with termination disabled.
>> setting the termination to the proper type in the Adaptec BIOS has been
>> known to solve a few problems. That's what this message is all about.
> ^^^^^
> Where is this knowledge going? Some time ago, it used to be in the
>various HOWTO documents, but they are pretty out of date by now
>(Hardware, Kernel, SCSI) or don't exist at all (SMP). For
>troubleshooting, we need checklists of known problems!
> ============================
>(Otherwise, the mailing lists will be flooded by FAQs.)
I guess that those of us who know the most of this information are too busy
with other things to write it all down, so we respond to specific emails on
the lists instead (it's somewhat easier to write a helpful hint to someone
when you know the problem and conditions than it is to write a generic help
document that covers all possible circumstances).
>> a difference. However, I haven't even so much as compile tested it, so I
>> make no gaurantees at this point in time. Let me know if you'd like an
>> alpha/beta patch for this problem :)
> ^^^^^^^^^^^^^^^^
> I am going to release one of the `cured' machines for SMP production,
>but can afford to reserve one for occasional tests etc. So, if you
>have a patch, I will certainly test it.
For the curious, I was referring to my current work on the aic7895 support,
which required overall about a two thirds re-write of the driver. So far, I
now have it compiling and am starting the functionality tests.....we'll see
how that goes.
----------------------------------
E-Mail: Doug Ledford <dledford@dialnet.net>
Date: 13-Jan-98
Time: 19:45:15
----------------------------------