Re: Bug in pg_dump/restore -o

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Philip Warner <pjw(at)rhyme(dot)com(dot)au>
Cc: Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Bug in pg_dump/restore -o
Date: 2002-01-18 04:24:29
Message-ID: 4002.1011327869@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Philip Warner <pjw(at)rhyme(dot)com(dot)au> writes:
> A sigle TOC entry, with both CREATE and COPY run at the start no matter
> what kind of dump/restore, seems like the solution. I'll look into it.

Okay. Keep in mind that given where we are in the release cycle, any
patch applied now has got to be minimal and easily checked. If what
you are thinking of involves any extensive changes, it'd probably be
better to let the problem go till 7.3, rather than risk introducing
worse breakage in 7.2. Especially since this problem isn't that major
in the big scheme of things (if it were, it'd have been found sooner).

One stopgap answer Bruce and I discussed on the phone was to temporarily
add a couple lines of code to pg_dump to reject the combination of -o
with -Fc or -Ft. That would prevent people from shooting themselves in
the foot in 7.2 until a proper solution can be devised for 7.3.

regards, tom lane

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2002-01-18 04:26:11 Re: Bug in pg_dump/restore -o
Previous Message Bruce Momjian 2002-01-18 04:20:01 Re: tuptoaster.c must *not* use SnapshotAny