From: Sergei Golubchik
Date: October 8 2009 8:36am
Subject: Re: bzr commit into mysql-5.1-bugteam branch (svoj:3146) Bug#47073
List-Archive: http://lists.mysql.com/commits/86097
Message-Id: <20091008083617.GA10968@janus.mylan>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit

Hi, Sergey!

On Oct 08, Sergey Vojtovich wrote:
> On Wed, Oct 07, 2009 at 09:38:27PM +0200, Sergei Golubchik wrote:
> 
> > > > Shouldn't you remove Ingo's fix for bug#25289 ?
> > > > Looks like it becomes a dead code after your changes.
> > > I'd leave this for safety. There is still a tiny chance that we fail to
> > > re-open data file.
> > 
> > No, I don't like it.
> > How do you see it could happen ?
> > If it cannot - add an assert, if you like, but not the code to hide a
> > possible bug.
> There already is an assert, but somewhat deeper. We may fail to open data
> file when it is e.g. removed during repair op.
> 
> OTOH I agree that leaving fix for BUG#25289 around is not a clean way.
> What do you think about something like this:
>   if (mi_open_data_file())
>     sort_param.retry_repair= 0;

That's fine.

And could you also add an assert to merge_many_buff() to show
that "t_file2 is always WRITE cache" ? Like

  DBUG_ASSERT(t_file2.type == WRITE_CACHE);

ok to push with these two changes
 
Regards / Mit vielen Grüßen,
Sergei

-- 
   __  ___     ___ ____  __
  /  |/  /_ __/ __/ __ \/ /   Sergei Golubchik <serg@stripped>
 / /|_/ / // /\ \/ /_/ / /__  Principal Software Engineer/Server Architect
/_/  /_/\_, /___/\___\_\___/  Sun Microsystems GmbH, HRB München 161028
       <___/                  Sonnenallee 1, 85551 Kirchheim-Heimstetten
Geschäftsführer: Thomas Schroeder, Wolfgang Engels, Wolf Frenkel
Vorsitzender des Aufsichtsrates: Martin Häring