On Wed, Jun 2, 2010 at 2:27 PM, Simon Riggs <simon@2ndquadrant.com> wrote:
> Syncing two servers in replication is common practice, as has been
> explained here; I'm still surprised people think otherwise. Measuring
> the time between two servers is the very purpose of the patch, so the
> synchronisation is not a design flaw, it is its raison d'etre.
I think the purpose of the patch should not be to measure the time
difference between servers, but rather the replication delay. While I
don't necessarily agree with Tom's statement that this is must-fix, I
do agree that it would be nicer if we could avoid depending on time
sync. Yeah, I keep my servers time synced, too. But, shit happens.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise Postgres Company