Re: pg_upgrade < 9.3 -> >=9.3 misses a step around multixacts

On Fri, May 30, 2014 at 05:25:26PM -0400, Bruce Momjian wrote:
> On Fri, May 30, 2014 at 11:03:53PM +0200, Andres Freund wrote:
> > On 2014-05-30 16:35:15 -0400, Bruce Momjian wrote:
> > > On Fri, May 30, 2014 at 05:13:17PM +0200, Andres Freund wrote:
> > > > It'd need to be clusters that used more multixacts than fit onto two
> > > > slru offsets/ segments in < 9.2. Otherwise things will probably just
> > > > continue to work because there's no hole.
> > >
> > > Do you mean <= 9.2 here for the old cluster?
> >
> > yes.
> >
> > > Multi-xacts were added in 9.3:
> >
> > That's not really correct. They were added in 8.2 or something...
> >
> > >         commit 0ac5ad5134f2769ccbaefec73844f8504c4d6182
> > >         Author: Alvaro Herrera <alvherre@alvh.no-ip.org>
> > >         Date:   Wed Jan 23 12:04:59 2013 -0300
> > >
> > >             Improve concurrency of foreign key locking
> >
> > That just expanded the usage to also be able to track updated rows.
>
> I guess I meant the new multixacts file format was in 9.3.

It appears this item is on hold until Alvaro can comment.  I can
probably find the impact and correct the problem for previous upgrades,
but it might be 1-2 months until I can get to it.  The fix for future
upgrades is simple.

--
  Bruce Momjian  <bruce@momjian.us>        http://momjian.us
  EnterpriseDB                             http://enterprisedb.com

  + Everyone has their own god. +