On Mon, 2010-10-18 at 17:51 -0400, Robert Haas wrote:
> On Mon, Oct 18, 2010 at 2:07 PM, Jeff Davis <pgsql@j-davis.com> wrote:
> > On Mon, 2010-10-18 at 17:02 +0900, Fujii Masao wrote:
> >> Yep, to automatically delete backup_label and continue recovery seems to be
> >> dangerous. How about just emitting FATAL error when neither restore_command
> >> nor primary_conninfo is supplied and backup_label exists? This seems to be
> >> simpler than your proposed patch (i.e., check whether REDO location exists).
> >>
> >
> > Either is fine with me.
> >
> > Do users have any expectation that they can restore a backup without
> > using recovery.conf by merely having the WAL segments in pg_xlog?
>
> I would expect that to work.
>
If that's the expectation, I believe my original fix has less impact.
It's essentially a sanity check to make sure that the first segment
needed is available before proceeding.
Regards,
Jeff Davis