Re: allow_in_place_tablespaces vs. pg_basebackup

On Thu, Mar 9, 2023 at 2:58 PM Kyotaro Horiguchi
<horikyota(dot)ntt(at)gmail(dot)com> wrote:
> At Wed, 8 Mar 2023 16:30:05 -0500, Robert Haas <robertmhaas(at)gmail(dot)com> wrote in
> > I'm not sure how messy it's going to be to clean this up. I think each
> > tablespace really needs to go into a separate ${TSOID}.tar file,
> > because we've got tons of code both on the server side and in
> > pg_basebackup that assumes that's how things work, and having in-place
> > tablespaces be a rare exception to that rule seems really unappealing.
> > This of course also implies that files in an in-place tablespace must
> > not go missing from the backup altogether.
>
> The doc clearly describes the purpose of in-place tablespaces and the
> potential confusion they can cause for backup tools not excluding
> pg_basebackup. Does this not provide sufficient information?
>
> https://www.postgresql.org/docs/devel/runtime-config-developer.html

Yeah. We knew that this didn't work (was discussed in a couple of
other threads), but you might be right that an error would be better
for now. It's absolutely not a user facing mode of operation, it was
intended just for the replication tests. Clearly it might be useful
for testing purposes in the backup area too, which is probably how
Robert got here. I will think about what changes would be needed as I
am looking at backup code currently, but I'm not sure when I'll be
able to post a patch so don't let that stop anyone else who sees how
to get it working...