From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Karsten Hilbert <Karsten(dot)Hilbert(at)gmx(dot)net> |
Cc: | pgsql-general <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: pg_upgrade 9.4 -> 9.5 with pg_trgm fails for me |
Date: | 2016-01-08 16:23:21 |
Message-ID: | 2530.1452270201@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Karsten Hilbert <Karsten(dot)Hilbert(at)gmx(dot)net> writes:
> On Fri, Jan 08, 2016 at 11:12:09AM -0500, Tom Lane wrote:
>> Hm. I just tried running a pg_upgrade here on a 9.4 database containing
>> pg_trgm 1.1, and didn't see any particular problem, so there's some
>> additional factor needed to cause your result. Hard to tell what.
>> Can you think of anything unusual about the history of your installation?
> No, other than that that cluster has been upgraded all the
> way from, I think, 8.4 over several Debian releases ;)
A suggestion is to run the pg_upgrade with -r switch, which will leave a
litter of files in your working directory. Some of them will be named
like pg_upgrade_dump_NNN.custom and should be custom-format, schema-only
dumps of your 9.4 installation's databases. If you'd be willing to send
those to me off-list, maybe I could figure out what's happening.
It occurs to me that this might actually be related to the issue you
saw in "dump" mode --- if there's some unresolved circular dependency,
it could cause pg_dump to dump things in an unexpected order, which
could possibly explain the message we're seeing. But that's just a
guess.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2016-01-08 16:28:33 | Re: pg_upgrade 9.4 -> 9.5 with pg_trgm fails for me |
Previous Message | Karsten Hilbert | 2016-01-08 16:15:22 | Re: pg_upgrade 9.4 -> 9.5 with pg_trgm fails for me |