On Thu, May 17, 2012 16:10, Ants Aasma wrote:
> On Thu, May 17, 2012 at 4:53 PM, Erik Rijkers <er@xs4all.nl> wrote:
>> The count(*) was done in the way that I showed, i.e. *after* psql had exited. My understanding
>> is
>> that, with synchronous replication 'on' and configured properly, psql could only return *after*
>> the sync-replica had the data safely on disk. Either my understanding is not correct or there
>> is
>> a bug in postgres sync-rep.
>
> Commit can only return when sync-replica has the data safely on disk,
> but this doesn't mean that it's visible yet.
>
> The sequence of events is in dot notation:
> commit_command -> master_wal_sync -> replica_wal_sync ->
> master_commit_visible -> commit_response
> replica_wal_sync -> replica_replay_wal -> replica_commit_visible
>
> If you issue a select on the replica after getting a commit response
> from master you can see that the query getting a snapshot races with
> replay of the commit record.
>
Ah yes, that makes sense. I hadn't thought of that.
Thank you for that explanation.