|From:||Bruce Momjian <bruce(at)momjian(dot)us>|
|To:||Hyongtae Lim <htlim(at)bitnine(dot)co(dot)kr>|
|Cc:||pgsql-hackers(at)postgresql(dot)org, Marti Raudsepp <marti(at)juffo(dot)org>|
|Subject:||Re: Re: [PATCH] pg_upgrade fails when postgres/template1 isn't in default tablespace|
|Views:||Raw Message | Whole Thread | Download mbox|
On Thu, Sep 10, 2015 at 09:59:41PM -0400, Bruce Momjian wrote:
> On Fri, Sep 11, 2015 at 12:41:09AM +0000, Hyongtae Lim wrote:
> > The following review has been posted through the commitfest application:
> > make installcheck-world: tested, passed
> > Implements feature: tested, passed
> > Spec compliant: tested, passed
> > Documentation: not tested
> > This looks good to me.
> > I've done some tests(pg_dumpall & restore with sql, upgrade test(8.4, 9.x, ...)) and there was no problems.
> > so I'm marking this as ready for committer.
> Thanks. I am going to commit it in the next 24 hours. Thanks.
Patch applied and marked as committed in the commitfest app.
One odd thing I found in exhaustive testing is that pre-9.3, we use a
line starting with \connect to identify when the pg_dumpall globals end
and the per-database information begins. (In 9.3+, we do per-database
dumps to allow parallel dump/restore.) I modified the patch for pre-9.3
to prefix a space for \connect lines when we are jumping in and out of
the database so we can alter its tablespace.
+ Everyone has their own god. +
|Next Message||Alvaro Herrera||2015-09-11 21:34:51||Re: [PROPOSAL] VACUUM Progress Checker.|
|Previous Message||Rahila Syed||2015-09-11 20:10:17||Re: [PROPOSAL] VACUUM Progress Checker.|