2000-11-29 11:11:21

by Jean-Luc Fontaine

[permalink] [raw]
Subject: 2.4 test10 error reading from HP colorado 7/14 Gb tape

Please CC me, thank you.

I can write just fine to the tape, but reading gives me the following
error:

# tar -tvf /dev/ht0
ide-tape: ht0: I/O error, pc = 8, key = 5, asc = 2c, ascq = 0

A search on deja.com shows that I am not the only one to have
experienced this and that it did not occur with previous versions of
ide-tape.c. The same error occurs with a non-modular kernel build.

The machine is a i386 redhat 7.0, kernel compiled with egcs 1.1.2.

--
Jean-Luc Fontaine mailto:[email protected] http://www.winealley.com




2000-11-29 12:26:18

by Eckhard Jokisch

[permalink] [raw]
Subject: Re: 2.4 test10 error reading from HP colorado 7/14 Gb tape

>
> # tar -tvf /dev/ht0
> ide-tape: ht0: I/O error, pc = 8, key = 5, asc = 2c, ascq = 0
>
> A search on deja.com shows that I am not the only one to have
> experienced this and that it did not occur with previous versions of
> ide-tape.c. The same error occurs with a non-modular kernel build.
>
I experience the same with OnStream DI30 - but these errors occure in 2.2.17 as
well. In my oppinion they show up in fewer cases with 2.4-test10.
The error is not quite"stable" because it occures on different tape positions
(filenames) when I retension the tape three aor four times befor writing to it.

I checked the tapes on a Windows machine and they seem to be o.k..

Eckhard Jokisch

2000-11-29 17:17:23

by Mike Dresser

[permalink] [raw]
Subject: Re: 2.4 test10 error reading from HP colorado 7/14 Gb tape

I reported this a few weeks ago or so, it seems that HP 7/14's are not exactly
standard. First is the proprietory tape size. Second is that the drive doesn't
support locking the tape in, but reports it as possible. At least, that's what i
gather from Jens's posting.

To be honest, I'm not sure the drives are worth fixing <grin> Unreliable media,
unreliable tape drives. I've got two dead out of 10, in under a year. And half a
dozen tapes or so. Plus, under 2.2.17+ide, the tape drive wouldn't restore tapes.
Something weird about the tape drive again. Had to boot up under plain 2.2.17 to
read my tapes.

Eckhard Jokisch wrote:

> >
> > # tar -tvf /dev/ht0
> > ide-tape: ht0: I/O error, pc = 8, key = 5, asc = 2c, ascq = 0
> >
> > A search on deja.com shows that I am not the only one to have
> > experienced this and that it did not occur with previous versions of
> > ide-tape.c. The same error occurs with a non-modular kernel build.
> >
> I experience the same with OnStream DI30 - but these errors occure in 2.2.17 as
> well. In my oppinion they show up in fewer cases with 2.4-test10.
> The error is not quite"stable" because it occures on different tape positions
> (filenames) when I retension the tape three aor four times befor writing to it.
>
> I checked the tapes on a Windows machine and they seem to be o.k..
>
> Eckhard Jokisch
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> Please read the FAQ at http://www.tux.org/lkml/

2000-11-29 20:19:26

by Eckhard Jokisch

[permalink] [raw]
Subject: Re: 2.4 test10 error reading from HP colorado 7/14 Gb tape

Mike Dresser wrote:
> I reported this a few weeks ago or so, it seems that HP 7/14's are not exactly
> standard. First is the proprietory tape size. Second is that the drive doesn't
> support locking the tape in, but reports it as possible. At least, that's what i
> gather from Jens's posting.
>

I'm very much interested in what is the cause for this error.
> > > # tar -tvf /dev/ht0
> > > ide-tape: ht0: I/O error, pc = 8, key = 5, asc = 2c, ascq = 0
> > >
In my case the driver tries some error-recovery three or four times before it
gives up.
On of the problems is - that these errors occure after having written several
GB to the tape - and that takes some hours. It also happend that with a tar
-cvM the first tape was ok but the second was unreadeable.

> > I experience the same with OnStream DI30 - but these errors occure in 2.2.17 as
> > well. In my oppinion they show up in fewer cases with 2.4-test10.
> > The error is not quite"stable" because it occures on different tape positions
> > (filenames) when I retension the tape three aor four times befor writing to it.
> >
> > I checked the tapes on a Windows machine and they seem to be o.k..
> >
> > Eckhard Jokisch
> > -