Jeff Davis wrote:
> Synchronous replication (to me) means that the data has been written to
> permanent storage on all masters and all slaves before any master or
> slave reports a successful COMMIT. Are you suggesting that you ship the
> WAL over the network, wait for it to be written to the slave, and then
> report a successful commit?
I'm not suggesting doing replication like that. But I think that's what
people mean when they propose that "multi-master synchronous replication
would be fairly trivial to implement with 2PC and wal-shipping".
This is more or less trivial to implement, yes. But it won't scale.
Regards
Markus