Re: Regression test PANICs with master-standby setup on same machine

Hello.

At Wed, 24 Apr 2019 13:23:04 +0900 (Tokyo Standard Time), Kyotaro HORIGUCHI <horiguchi(dot)kyotaro(at)lab(dot)ntt(dot)co(dot)jp> wrote in <20190424(dot)132304(dot)40676137(dot)horiguchi(dot)kyotaro(at)lab(dot)ntt(dot)co(dot)jp>
> > > with a check that forces relative paths to be outside of PGDATA (baring
> > > symlinks). As far as I can tell convert_and_check_filename() would check
> > > just about the opposite.
>
> We need to adjust relative path between PGDATA-based and
> pg_tblspc based. The attached first patch does that.
>
> - I'm not sure it is OK to use getcwd this way. Another issue
> here is is_in_data_directory canonicalizes DataDir
> on-the-fly. It is needed when DataDir contains '/./' or such. I
> think the canonicalization should be done far earlier.
>
> The second attached is TAP change to support tablespaces using
> relative tablespaces.
>
> - This is tentative or sample. I'll visit the current discussion thread.
>
> The third is test for this issue.
>
> - Tablespace handling gets easier.
>
> The fourth is the fix for the issue here.
>
> - Not all possible simliar issue is not checked.

This is new version. Adjusted pg_basebackup's behavior to allow
relative mappings. But..

This is apparently out of a bug fix. What should I do with it?

Should we applying only 0004 (after further checking) or
something as bug fix, then register the rest for v13?

regards.

--
Kyotaro Horiguchi
NTT Open Source Software Center