List:Replication« Previous MessageNext Message »
From:Mark Callaghan Date:May 9 2008 8:57pm
Subject:Re: question about encoding of auto-commit events in the binlog
View as plain text  
That is related, but http://bugs.mysql.com/bug.php?id=26395 has the change
to always use BEGIN/COMMIT events in the binlog.

On Wed, May 7, 2008 at 7:45 PM, Eric Bergen <eric@stripped> wrote:

> Hi Mark,
>
> Are you referring to the fix for http://bugs.mysql.com/bug.php?id=29288
>
> -Eric
>
>
> On Wed, May 7, 2008 at 2:39 PM, Mark Callaghan <mcallaghan@stripped>
> wrote:
> > I thought I read a CL for a bug fix that stated that auto-commit events
> >  would be encoded differently as of a new MySQL 5.0 release. The new
> encoding
> >  is to always wrap the SQL statement with binlog events for BEGIN and
> COMMIT.
> >  Is that the case, and if so, in which release does this change occur?
> >
> >  --
> >  Mark Callaghan
> >  mcallaghan@stripped
> >
>
>
>
> --
> high performance mysql consulting.
> http://provenscaling.com
>



-- 
Mark Callaghan
mcallaghan@stripped

Thread
question about encoding of auto-commit events in the binlogMark Callaghan7 May 2008
  • Re: question about encoding of auto-commit events in the binlogMark Callaghan8 May 2008
  • Re: question about encoding of auto-commit events in the binlogEric Bergen8 May 2008
    • Re: question about encoding of auto-commit events in the binlogMark Callaghan9 May 2008