Re: pg_upgrade - Mailing list pgsql-bugs

From Tomasz Szypowski
Subject Re: pg_upgrade
Date
Msg-id CACmJi2J2stMNKuBm=sLrjhhJh68AWEoDCnAM5UJ5rj1N=wBKTw@mail.gmail.com
Whole thread Raw
In response to Re: pg_upgrade  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-bugs
As far as I remember pg_upgrade is from 11.2 rest is from 9.5. This is due to the fact, that the version is secured, only md5, md5 hashed and so on. I compared the code and didn’t see much difference in pg_upgrade core, but give me some days and I will test it using binaries from 11.2

Regards
Thomas

W dniu pon., 18.03.2019 o 23:37 Tom Lane <tgl@sss.pgh.pa.us> napisał(a):
Tomasz Szypowski <tomasz.szypowski@gmail.com> writes:
> So what set it to false?

I was hoping you'd tell me ;-).  If it's not false in the old cluster,
though, that theory is all wet.

I wonder if you're somehow using the wrong version of pg_upgrade or
pg_dump.  There are cross-checks about that in pg_upgrade, but it
looks like they only check the major version number --- if we'd changed
anything about this in a minor release (which I think we did), it might
be possible to get burnt if you were using pg_upgrade or pg_dump from a
prior minor release.  What are all the versions involved, exactly?

                        regards, tom lane
--
pozdrawiam
Tomek

pgsql-bugs by date:

Previous
From: Tom Lane
Date:
Subject: Re: pg_upgrade
Next
From: Tom Lane
Date:
Subject: Re: pg_upgrade