(Fwd) Re: SCSI tape data loss

[Dan Langille]

On 27 Aug 2003 at 11:06, Nate Lawson wrote:

> On Wed, 27 Aug 2003, Kern Sibbald wrote:
> > Many thanks for testing this ...
> >
> > It seems to me that your tests, clearly indicate that
> > there is a problem even though you had no data loss.
> >
> > When you ran without -pthread, the status
> > received by the program was a 0 with 141776
> > blocks written. This is correct.
> >
> > When you ran with -pthread, the status
> > received by the program was a -1 with 142879
> > blocks written. This is "not correct".
> >
> > To me, that shows very clearly that with -pthread
> > the 0 status was lost and more blocks were written.
> > In fact, in this case so many blocks were written
> > that the tape was not properly terminated with
> > an EOF (actually two EOF marks).
> 
> Here is a response I got by forwarding this to the pthreads maintainer:
> > A return status of 0 from write is not interpreted as an End-Of-Tape.
> > The threads library isn't smart enough to know that the file
> > is a tape device and that a 0 status should break it out of the
> > loop.  Thus, it continues writing.
> >
> > Use libkse :-)
> >
> > --
> > Dan Eischen

Nate: thanks for getting in touch with him.

It is interesting to note that the code works OK on Linux and 
Solaris.  Why is FreeBSD different in this case?

Kern: I can't comment on libkse.  I don't know it and I don't know 
what effect it would have on Bacula.

-- 
Dan Langille : http://www.langille.org/