From: Tom Lane <tgl@sss.pgh.pa.us>
To: Amit Langote <amitlangote09@gmail.com>
cc: fedor_erastov@mail.ru,
        PostgreSQL mailing lists <pgsql-bugs@lists.postgresql.org>,
        Andres Freund <andres@anarazel.de>,
        Andrew Dunstan <andrew@dunslane.net>
Subject: Re: BUG #16644: null value for defaults in OLD variable for trigger
In-reply-to: 
 <CA+HiwqH7ys=WmnbGm+wgQwt0WzvQ-R-Qn4vdE01Q_QsX7cyFsw@mail.gmail.com>
References: <16644-5da7ef98a7ac4545@postgresql.org>
 <802555.1601432494@sss.pgh.pa.us>
 <CA+HiwqH7ys=WmnbGm+wgQwt0WzvQ-R-Qn4vdE01Q_QsX7cyFsw@mail.gmail.com>
Comments: In-reply-to Amit Langote <amitlangote09@gmail.com>
	message dated "Sat, 24 Oct 2020 00:36:56 +0900"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <207141.1603468329.1@sss.pgh.pa.us>
Date: Fri, 23 Oct 2020 11:52:09 -0400
Message-ID: <207142.1603468329@sss.pgh.pa.us>
Precedence: bulk

Amit Langote <amitlangote09@gmail.com> writes:
> On Wed, Sep 30, 2020 at 11:21 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> I've not tried to find where is the difference between 11 and
>> 12 that makes it fail or not fail.

> In v11, GetTupleForTrigger() expands the HeapTuple, with this:

>     /*
>      * While this is not necessary anymore after 297d627e, as a defense
>      * against C code that has not recompiled for minor releases after the
>      * fix, continue to expand the tuple.
>      */
>     if (HeapTupleHeaderGetNatts(tuple.t_data) < relation->rd_att->natts)
>         result = heap_expand_tuple(&tuple, relation->rd_att);
>     else
>         result = heap_copytuple(&tuple);
>     ReleaseBuffer(buffer);

Ah, good sleuthing.

>> I think we can band-aid this immediate problem by forcing
>> trigger.c to materialize the "old" tuples it fetches from disk
>> (or whatever needs to be done to substitute missing values into
>> them).

> Maybe something like the attached?

Probably needs some attention to memory management (e.g.,
should_free_trig) but I'm okay with doing this as a short-term
fix.  I remain worried about similar instances elsewhere, but
I have no good ideas about how to find them.

			regards, tom lane