On Wed, Jul 27, 2016 at 1:47 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> I looked into the problem reported in bug #14265,
> attached is a proposed patch that fixes this bug
> this would require an initdb because it changes the
> representation of stored rules for cases like this,
So pg_upgrade would not work at all for the version this goes into,
or do you see us testing for such cases and allowing pg_upgrade if
none are found? Or do you see a way to allow pg_upgrade in the
face of such stored rules?
--
Kevin Grittner
EDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company