Le 28 déc. 2009 à 22:46, Andres Freund a écrit :
>> <mad proposal time>
>> I'd be in favor of considering how to get pgbouncer into -core, and now
>> that we have Hot Standby maybe implement a mode in which as soon as a
>> "real" XID is needed, or maybe upon receiving start transaction read write
>> command, the connection is handled transparently to the master. </>
> Thats not as easy as it sounds - the master may not have all data needed by
> the snapshot on the slave anymore.
I suppose that if it was easy some patch would already be around for next commit fest? :)
Seriously, your point is why I'd be tempted to only consider getting to the master at transaction starting time. That
isbefore any snapshot is taken.
Regards,
--
dim