[7275] in linux-scsi channel archive
Re: aic7xxx 5.1.19/3.2.4: "Performing Domain validation"
daemon@ATHENA.MIT.EDU (Ulrich Windl)
Wed Sep 29 02:29:18 1999
From: "Ulrich Windl" <ulrich.windl@rz.uni-regensburg.de>
To: Doug Ledford <dledford@redhat.com>
Date: Wed, 29 Sep 1999 08:26:50 +0200
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Cc: linux-scsi@vger.rutgers.edu
In-reply-to: <37F0F7D9.FAD05D36@redhat.com>
Message-ID: <2029832A4B72@rkdvmks1.ngate.uni-regensburg.de>
On 28 Sep 99, at 13:16, Doug Ledford wrote:
> Ulrich Windl wrote:
> >
> > Hello,
> >
> > I noticed new messages when accessing my HP35480A DAT in Linux-
> > 2.2.12. I have a AHA2940. The messages are
> > (scsi0:2:0) Performing Domain validation
> > (scsi0:2:0) Successfully completed Domain verification
> >
> > ID 2 is my DAT. I should note that I did not change the tape for the
> > message to appear; I just rewinded the tape and read it. Is that a
> > bug in the driver?
>
> No, it just means you have the driver set in verbose mode. When the driver
> gets a CHECK_CONDITION return from a device, we do level 1 domain validation
> just to make sure it isn't a cable quality issue. That gives the driver some
> rudimentary ability to reduce transfer speeds to something that doesn't cause
> bus problems if that is the case. In your case, it was a check condition
> triggered by the rewinding (I would assume anyway) and we did a domain
> validation and everything was fine so we go on with life.
I read some comment in the driver about a change in some mode or
status page that could trigger it. I know that the HP DAT tapes keep
various performance and usage statistics (e.g. corrected read errors,
tape loads, etc.) in some mode pages, but I only have a tool for OS/2
to display those. I have no idea how to get the specs from HP...
Yes, I have the driver in verbose mode, but the question is why a
domain validation is started that frequently. Isn't a CHECK CONDITION
a kind of exceptional event? I can hardly imagine tha trewinding
caused it.
Regards,
Ulrich
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
