Re: Backup history file should be replicated in Streaming Replication?

On Tue, 2009-12-22 at 22:46 +0200, Heikki Linnakangas wrote:
> Simon Riggs wrote:
> > On Wed, 2009-12-23 at 04:15 +0900, Fujii Masao wrote:
> >> On Wed, Dec 23, 2009 at 4:09 AM, Simon Riggs <simon(at)2ndquadrant(dot)com> wrote:
> >>> It's needed to find the place where the backup stopped, so it defines
> >>> the safe stopping point. We could easily pass that info via WAL, when
> >>> streaming. It doesn't actually matter until we try to failover.
> >> Right. And, it's also needed to cooperate with HS which begins accepting
> >> read-only queries after a recovery reaches that safe stopping point.
> >
> > Agreed, hence my interest!
>
> Yeah, that's a great idea.
>
> I was just having a chat with Magnus this morning, and he asked if the
> current patch already provides or if it would be possible to write a
> stand-alone utility to connect to a master and stream WAL files to an
> archive directory, without setting up a full-blown standby instance. We
> came to the conclusion that backup history files wouldn't be copied as
> the patch stands, because the standby has to specifically request them.

There isn't any need to write that utility. Read my post about 2-phase
backup and you'll see we are a couple of lines of code away from that.

--
Simon Riggs www.2ndQuadrant.com