Re: Streaming basebackups vs pg_stat_tmp

On Sat, Oct 29, 2016 at 4:12 PM, Michael Paquier <michael(dot)paquier(at)gmail(dot)com>
wrote:

> On Fri, Oct 28, 2016 at 9:57 PM, David Steele <david(at)pgmasters(dot)net> wrote:
> > On 10/28/16 3:49 PM, Magnus Hagander wrote:
> > The change from 10 to 11 increases the tests that are skipped on Windows,
> > which is necessary because one extra symlink test is added.
> >
> > I think you need:
> >
> > [...]
> >
> > The rest of the tests are for exclusions.
>
> Indeed, giving the attached for REL9_6_STABLE. You could as well have
> a test for pg_stat_tmp but honestly that's not worth it. One thing I
> have noticed is that the patch does not use _tarWriteDir() for
> pg_xlog. I think it should even if that's not addressing directly a
> bug...
>

Applied and backported, thanks. Backported to 9.4, as this is where that
exclusion code appeared.

I did not backport the tests, as we don't have the $node stuff available in
9.5 and earlier.

--
Magnus Hagander
Me: http://www.hagander.net/
Work: http://www.redpill-linpro.com/