Re: client_min_messages in dumps? - Mailing list pgsql-hackers

From Tom Lane
Subject Re: client_min_messages in dumps?
Date
Msg-id 24850.1088490444@sss.pgh.pa.us
Whole thread Raw
In response to client_min_messages in dumps?  (Christopher Kings-Lynne <chriskl@familyhealth.com.au>)
Responses Re: client_min_messages in dumps?  (Dennis Bjorklund <db@zigo.dhs.org>)
Re: client_min_messages in dumps?  (Bruce Momjian <pgman@candle.pha.pa.us>)
List pgsql-hackers
Christopher Kings-Lynne <chriskl@familyhealth.com.au> writes:
> Would anyone else think that dumping:
>   SET client_min_messages TO warning;
> In pg_dumps would be cool?

> It would mean that while restoring a dump you can actually see the wood 
> for the trees when trying to see ERRORs and WARNINGs the the massive 
> spam of messages you get while restoring.

Another answer is to get rid of some of our chattier NOTICEs, or at
least downgrade 'em to DEBUG1 or thereabouts.  Does anyone really still
think that
NOTICE:  CREATE TABLE / PRIMARY KEY will create implicit index "foo_pkey" for table "foo"
is conveying any useful information?  It's not like you couldn't find
out the name of the index from "\d foo".

I think this is a better answer than having pg_dump try to force the
message level, since that would get in the way of seeing the messages
when you needed to do so for debugging reasons.

(cc'ing to Bruce who has always been the most vocal defender of those
notices... now where did I put my flameproof longjohns?)
        regards, tom lane


pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: bug in pg_dump ALTER DATABASE
Next
From: Darko Prenosil
Date:
Subject: Re: improper call to spi_printtup ???