Date: Thu, 6 Apr 1995 16:10:26 -0700 (PDT) From: julian@TFS.COM (Julian Elischer) To: tst@condor.sys.swt.edu (Thomas S. Traylor) Cc: freebsd-hardware@FreeBSD.org Subject: Re: Conner CTMS 3200 SCSI Tape Drive Message-ID: <m0rx0h1-00040XC@TFS.COM> In-Reply-To: <Pine.OSF.3.91.950406171145.24339B-100000@condor.sys.swt.edu> from "Thomas S. Traylor" at Apr 6, 95 05:15:28 pm
next in thread | previous in thread | raw e-mail | index | archive | help
>
> Dump doesn't have a problem writing to the tape but restore is unable to
> read the tape. (see below)
>
> vulture# dump 0f /dev/rst0 /
> DUMP: Date of this level 0 dump: Thu Apr 6 16:01:46 1995
> DUMP: Date of last level 0 dump: the epoch
> DUMP: Dumping /dev/rsd0a (/) to /dev/rst0
> DUMP: mapping (Pass I) [regular files]
> DUMP: mapping (Pass II) [directories]
> DUMP: estimated 13001 tape blocks on 0.33 tape(s).
> DUMP: dumping (Pass III) [directories]
> DUMP: dumping (Pass IV) [regular files]
> DUMP: DUMP: 13003 tape blocks on 1 volumes(s)
> DUMP: Closing /dev/rst0
> DUMP: DUMP IS DONE
> vulture# restore if /dev/rst0
> tape read error: Input/output error
> vulture#
This is a BUG in the present code..
you can get around it by telling restore to read in 10K records
(what dump used to write)
the case it's getting is a short read..
(read 30k, get 10k data, (that's all dump wrote on each record).
it turns out this is NOT an error, but the driver get's confused
and reports it as an error..
it's being fixed as we speak..
(this also affects DAT drives and others that by default use variable
block mode)....
>
> This is what is written to /var/log/messages:
>
> /kernel: st0(aha0:3:0): NO SENSE info:5800
^^^^
5 x 4096 + 2048
= 22k,
the difference between
the 10k written data
and the 32k request for data.
(we're working on it)
(what version are you running?)
julian
>
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?m0rx0h1-00040XC>