From: Tom Lane <tgl@sss.pgh.pa.us>
To: enderstd@gmail.com
cc: pgsql-bugs@lists.postgresql.org
Subject: Re: BUG #18130: \copy fails with "could not read block" or "page
 should be empty but not" errors due to triggers
In-reply-to: <18130-7a86a7356a75209d@postgresql.org>
References: <18130-7a86a7356a75209d@postgresql.org>
Comments: In-reply-to PG Bug reporting form <noreply@postgresql.org>
	message dated "Fri, 22 Sep 2023 11:39:12 -0000"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <262083.1695672850.1@sss.pgh.pa.us>
Date: Mon, 25 Sep 2023 16:14:10 -0400
Message-ID: <262084.1695672850@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/262084.1695672850%40sss.pgh.pa.us>
Precedence: bulk

PG Bug reporting form <noreply@postgresql.org> writes:
> I'm trying to copy a CSV file to a table and it errors out with different
> error messages every time:

Thank you for the report and reproducible test case!  It looks to me
like this is probably a bug in our new logic for extending tables
more than one block at a time, and the connection to triggers is
accidental.  I brought it up in a thread that was already discussing
the extension logic [1].

			regards, tom lane

[1] https://www.postgresql.org/message-id/257696.1695670946%40sss.pgh.pa.us