> On Wed, Nov 02, 2005 at 03:44:26PM -0800, Marc Munro wrote:
>> experts there may suggest a better solution. I have seen talk of
>> disabling the standard slony triggers to allow this sort of thing but
>> whether that is more or less nasty is questionable.
>
> FWIW, I don't think that's the question; it's more like whether it'd
> be merely horribly nasty or likely to break in unexpected and really
> painful ways. ;-) But the discussion around that surely should move
> to the Slony list.
It seems to me that lots of the "stuff" in Slony-I could be reapplied
to _try_ to create an asynchronous multimaster replication system.
A *major* addition would need to be some form of "conflicts queue."
That's the sort of thing they have in the analagous "O-word"
replication system.
What's a non-starter is to try to reshape the Slony-I project into
"async multimaster." That would get considerable push-back :-).
But if someone decided to "fork" their own *new* project, perhaps
starting based on one of the releases, that would an entirely
interesting idea.
--
output = reverse("moc.enworbbc" "@" "enworbbc")
http://cbbrowne.com/info/languages.html
To quote from a friend's conference talk: "they told me that their
network was physically secure, so I asked them `then what's with all
these do-not-leave-valuables-in-your-desk signs?'".
-- Henry Spencer