At Tue, 23 Apr 2019 14:53:28 +0900, Michael Paquier <michael@paquier.xyz> wrote in <20190423055328.GK2712@paquier.xyz>
> On Tue, Apr 23, 2019 at 01:33:39PM +0900, Kyotaro HORIGUCHI wrote:
> > I think this is rahter a testing or debugging feature. This can
> > be apply to all paths, so the variable might be "path_prefix" or
> > something more generic than tablespace_chroot.
> >
> > Does it make sense?
>
> A GUC which enforces object creation does not sound like a good idea
> to me, and what you propose would still bite back, for example two
> local nodes could use the same port, but a different Unix socket
> path.
It's not necessarily be port number, but I agree that it's not a
good idea. I prefer allowing relative paths for tablespaces.
regards.
--
Kyotaro Horiguchi
NTT Open Source Software Center