Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 11 Mar 1998 23:47:30 +0100
From:      Stefan Esser <se@FreeBSD.ORG>
To:        Josh Tiefenbach <josh@zipperup.org>, scsi@FreeBSD.ORG
Cc:        gibbs@narnia.plutotech.com, Stefan Esser <se@FreeBSD.ORG>
Subject:   Re: CAM Testimonials
Message-ID:  <19980311234730.18013@mi.uni-koeln.de>
In-Reply-To: <19980311132147.52937@sun.zipperup.org>; from Josh Tiefenbach on Wed, Mar 11, 1998 at 01:21:47PM -0500
References:  <19980311132147.52937@sun.zipperup.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On 1998-03-11 13:21 -0500, Josh Tiefenbach <josh@zipperup.org> wrote:
> I havent been able to get the latest CAM snap to work on my machine.
> 
> Specs:
> 
> -current kernel ca 3/10/98, userland bits from ~ 1/27/98.
> Asus SC200 (53c810 based card)
> Micropolis Stinger 4gig HD (MICROP 4743 S150)
> SCSI CD-ROM (MATSHITA CD-ROM CR-506 8S05)
> 
> The symptoms:
> 
> Kernel boots up to the point where the CAM code would start scanning the bus.
> At this point, I get the following messages:
> 
> Mar 10 23:12:00 asherah /kernel: ncr0: SCSI phase error fixup: CCB already
> dequeued (0xf085ac00)
> Mar 10 23:12:00 asherah /kernel: ncr0: timeout nccb=f085a800 (skip)
> Mar 10 23:12:00 asherah /kernel: ncr0: timeout nccb=f085c800 (skip)
> 
> Followed by a lot more ncr timeouts. I've left it for ~5 minutes, with no
> change.

Can't really tell what's wrong. Are these the only SCSI and
NCR messages ?

> Interestingly enough, when I boot with a `normal' kernel I get:
> 
> sd0 at scbus0 target 0 lun 0
> sd0: Direct-Access 
> sd0: 10.0 MB/s (100 ns, offset 8)
> 
> sd0: M_DISCONNECT received, but datapointer not saved:
>         data=1e1b4 save=3226b0 goal=3226d4.

This is most likely caused by the drive sending back less 
than 36 bytes as response to an INQUIRE command.

> I also get the odd
> 
> Mar 10 13:24:24 asherah /kernel: ncr0: timeout ccb=f07c3400 (skip)
> 
> >From time to time with the old code.

Shouldn't happen, but this message is not sufficient to 
tell what's wrong.

> I'm currently away from the machine right now, but if you want me to try any
> diagnostics, I can do it when I get home.

It would be interesting to know, which SCSI command fails.
I guess, that the NCR driver does not correctly interoperate
with the CAM layer, if a short read (less than 36 bytes) has
occured.

I'll try building a kernel with the latest CAM code and will
let you know, what I find.

Regards, STefan

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-scsi" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19980311234730.18013>