From: Tom Lane <tgl@sss.pgh.pa.us>
To: Robert Haas <robertmhaas@gmail.com>
cc: "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
Subject: Re: pgsql: Harden new test case against force_parallel_mode = regress.
In-reply-to: <CA+TgmoY-dsB-6GyF8q+V0T5+0Gbw0cT6TE7_8SsnESeUHZpnZQ@mail.gmail.com>
References: <E1pXri2-001aid-6W@gemulon.postgresql.org> <CA+TgmoY-dsB-6GyF8q+V0T5+0Gbw0cT6TE7_8SsnESeUHZpnZQ@mail.gmail.com>
Comments: In-reply-to Robert Haas <robertmhaas@gmail.com>
	message dated "Fri, 03 Mar 2023 11:16:28 -0500"
MIME-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-ID: <2478845.1677861476.1@sss.pgh.pa.us>
Content-Transfer-Encoding: quoted-printable
Date: Fri, 03 Mar 2023 11:37:56 -0500
Message-ID: <2478846.1677861476@sss.pgh.pa.us>
Archived-At: <https://www.postgresql.org/message-id/2478846.1677861476%40sss.pgh.pa.us>
Precedence: bulk

Robert Haas <robertmhaas@gmail.com> writes:
> On Thu, Mar 2, 2023 at 5:47=E2=80=AFPM Tom Lane <tgl@sss.pgh.pa.us> wrot=
e:
>> Per buildfarm: worker processes can't see a role created in
>> the current transaction.

> Now why would that happen? Surely the snapshot for each command is
> passed down from leader to worker, and the worker is not free to
> invent a snapshot from nothing.

The workers were failing at startup, eg (from [1]):

+ERROR:  role "regress_psql_user" does not exist
+CONTEXT:  while setting parameter "session_authorization" to "regress_psq=
l_user"

Maybe this says that worker startup needs to install the snapshot before
doing any catalog accesses?  Anyway, I'd be happy to revert this test
hack if you care to make the case work.

			regards, tom lane