On Wed, May 05, 2004 at 04:22:12PM -0700, Jeremy Zawodny wrote:
> On Wed, May 05, 2004 at 03:44:00PM -0400, Luis R. Rodriguez wrote:
> > On Wed, May 05, 2004 at 12:01:27PM -0700, Jeremy Zawodny wrote:
> > > On Wed, May 05, 2004 at 01:07:58PM -0400, Luis R. Rodriguez wrote:
> > > >
> > > > Hello,
> > > >
> > > > I am wondering if anyone is currently working on a two way replication
> > > > lock protocol. If so, what's the status. I understand there's little
> > > > gain in having two way replication but -- I am just curious.
> > >
> > > What exactly do you mean by two-way replication?
> >
> > Well I guess it would be a lock protocol for replication among master
> > servers, not just two :).
>
> What's the locking for?
Scenario:
A <--> B
A is master to B,
B is master to A.
Problem: Update goes to A, and another one goes to B. You have the
typical race condition if no lock is in place.
One way to go about this is that if an update goes to be B, first go and try to lock
A; if
successful, go ahead and writ eto B and A. Then unlock.
That is the type of locking I am talking about.
Luis
--
GnuPG Key fingerprint = 113F B290 C6D2 0251 4D84 A34A 6ADD 4937 E20A 525E
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
