Re: [HACKERS] pg_dump - segfault with -z option

From: Constantin Teodorescu <teo(at)flex(dot)ro>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: Karl Auer <auer(at)kom(dot)id(dot)ethz(dot)ch>, Oleg Bartunov <oleg(at)sai(dot)msu(dot)su>, PostgreSQL Hacker <pgsql-hackers(at)postgreSQL(dot)org>
Subject: Re: [HACKERS] pg_dump - segfault with -z option
Date: 1998-12-05 18:44:19
Message-ID: 36697F03.AECC4899@flex.ro
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Tom Lane wrote:
>
> I have found one cause of sometimes-but-not-always coredumps in
> pg_dump -z. Please apply the following patch and let me know whether
> things get better for you.

I have applied the patch and test it against the same database.
Now it's ok, no more core dumps, thanks a lot.

> The ACL code in pg_dump.c looks really butt-ugly ... I think I will
> rewrite the whole thing this afternoon ... but this one bugfix may be
> enough to get you going for now.

If you are going to rewrite some of the portions of pg_dump, I would
like to ask you to include that tiny patch that makes pg_dump write for
table structures just 1 field per line (inserting \n\t after every
field) in order to get a better look of the file. The current version is
giving a extremely long line for tables and it's somehow difficult to
watch or edit it.

All the best,

--
Constantin Teodorescu
FLEX Consulting Braila, ROMANIA

Browse pgsql-hackers by date

  From Date Subject
Next Message Oliver Elphick 1998-12-05 19:57:15 Parser bug (?)
Previous Message Oleg Broytmann 1998-12-05 16:40:45 Locale fix