Re: using index or check in ALTER TABLE SET NOT NULL

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Andres Freund <andres(at)anarazel(dot)de>
Cc: Robert Haas <robertmhaas(at)gmail(dot)com>, David Steele <david(at)pgmasters(dot)net>, David Rowley <david(dot)rowley(at)2ndquadrant(dot)com>, Sergei Kornilov <sk(at)zsrv(dot)org>, PostgreSQL mailing lists <pgsql-hackers(at)postgresql(dot)org>, Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org>, Stephen Frost <sfrost(at)snowman(dot)net>, Ildar Musin <i(dot)musin(at)postgrespro(dot)ru>
Subject: Re: using index or check in ALTER TABLE SET NOT NULL
Date: 2019-05-01 16:38:01
Message-ID: 25570.1556728681@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Andres Freund <andres(at)anarazel(dot)de> writes:
> There's still an open item "Debate INFO messages in ATTACH PARTITION and
> SET NOT NULL" for this thread. Is there anything further to be done? Or
> are we content with this for v12?

IIRC, that's based on my complaint that these messages have no business
being INFO level. I'm definitely not content with the current state.

I proposed that we didn't need those messages at all, which Robert pushed
back on, but I'd be willing to compromise by reducing them to NOTICE or
DEBUG level.

The actually intended use of INFO level is to mark messages that the user
actively asked for (via VERBOSE or some morally-equivalent option), which
is why that level has such high priority. If we had something like a
VERBOSE option for ALTER TABLE, it'd make plenty of sense to follow
VACUUM's precedent:

if (params->options & VACOPT_VERBOSE)
elevel = INFO;
else
elevel = DEBUG2;

It seems a bit late to be inventing such a thing for v12 though.

In the meantime, I'm not very content with random subsets of
ALTER TABLE acting as if they have been granted permission
to be VERBOSE. It's unfriendly, and it's inconsistent because
there are so many other expensive ALTER TABLE actions that don't
do this.

regards, tom lane

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Robert Haas 2019-05-01 16:39:13 Re: New vacuum option to do only freezing
Previous Message Robert Haas 2019-05-01 16:34:04 Re: pgsql: Compute XID horizon for page level index vacuum on primary.