Re: BUG #15183: pg_upgrade and pg_upgrade.exe both corrupt sequences when migrating 9.1 10.3

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: decalod85(at)gmail(dot)com
Cc: pgsql-bugs(at)lists(dot)postgresql(dot)org
Subject: Re: BUG #15183: pg_upgrade and pg_upgrade.exe both corrupt sequences when migrating 9.1 10.3
Date: 2018-05-02 14:47:09
Message-ID: 18109.1525272429@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

=?utf-8?q?PG_Bug_reporting_form?= <noreply(at)postgresql(dot)org> writes:
> On the most recent version, after the pg_upgrade ran, we were unable to
> insert records. Analysis showed that our normally very simple sequences had
> multiple dependencies attached to them (in the 10.3 database). The 9.1
> database only had a single dependency on the sequence. The pattern of which
> dependencies were added to which sequences was not discernible, but it was
> the same on both Windows and Linux.

You haven't really provided enough information to let anyone else
reproduce or investigate this. What do you mean by "multiple dependencies
attached"? How would that have broken your application? What exactly
is the database schema that pg_upgrade failed to reproduce accurately?

Ideally, what I'd like to have to avoid wasting time reverse-engineering
your problem is a SQL script that creates a 9.1 database that pg_upgrade
will then fail on.

regards, tom lane

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message anderson.hsu 2018-05-03 05:54:35 postgresql license question
Previous Message Tom Lane 2018-05-02 14:42:27 Re: power() function in Windows: "value out of range: underflow"