MIME-Version: 1.0
References: <CAJ7c6TOZxbXoyh0h-Fvz9smJoV8JGrG4Wi-T6F+tVYyrVGV2nw@mail.gmail.com>
 <CABUevEyZO1WnaxtChbjx4EMTtMtvNwy_ebOK1XY9YqQyDDdcfw@mail.gmail.com>
 <ZZ6h/enpsDFkpIUu@ip-10-97-1-34.eu-west-3.compute.internal>
 <CABUevEz74kTPWKjJSBb8KxQq0518p3eDxgLBQZtJKFGZWojvyw@mail.gmail.com>
 <ZZ6lzKuwzNTqVE9D@ip-10-97-1-34.eu-west-3.compute.internal>
 <CABUevEx8KfR=Bj8suZQ5TMouBDJ3A+byWTBXf1=vHc6DfMgHwQ@mail.gmail.com>
 <ZaAdeVa2+eN4wTzS@ip-10-97-1-34.eu-west-3.compute.internal>
 <CABUevEzYFpJbr_SYxh_XAX7FDiC4K1TC=dHjZydGY26V+FUsjQ@mail.gmail.com>
 <ZaUGPkTPphJhG0lO@ip-10-97-1-34.eu-west-3.compute.internal>
 <CABUevEyy6==7_pm8DH1GxKyOf7oo0Gf3Tq9ey9=9991d4+dkHQ@mail.gmail.com> <ZaoUkEzoroFwroGe@ip-10-97-1-34.eu-west-3.compute.internal>
In-Reply-To: <ZaoUkEzoroFwroGe@ip-10-97-1-34.eu-west-3.compute.internal>
From: Magnus Hagander <magnus@hagander.net>
Date: Fri, 16 Feb 2024 20:17:41 +0100
Message-ID: <CABUevEw9kE3kB-e85fe2uv0HODrpB_mchamJLhM3pX5=m0aRcQ@mail.gmail.com>
Subject: Re: System username in pg_stat_activity
To: Bertrand Drouvot <bertranddrouvot.pg@gmail.com>
Cc: Aleksander Alekseev <aleksander@timescale.com>, 
	PostgreSQL-development <pgsql-hackers@postgresql.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Archived-At: <https://www.postgresql.org/message-id/CABUevEw9kE3kB-e85fe2uv0HODrpB_mchamJLhM3pX5%3Dm0aRcQ%40mail.gmail.com>
Precedence: bulk

On Fri, Jan 19, 2024 at 7:20=E2=80=AFAM Bertrand Drouvot
<bertranddrouvot.pg@gmail.com> wrote:
>
> Hi,
>
> On Thu, Jan 18, 2024 at 04:01:33PM +0100, Magnus Hagander wrote:
> > On Mon, Jan 15, 2024 at 11:17=E2=80=AFAM Bertrand Drouvot
> > > Did you forget to share the new revision (aka v4)? I can only see the
> > > "reorder_parallel_worker_bestart.patch" attached.
> >
> > I did. Here it is, and also including that suggested docs fix as well
> > as a rebase on current master.
> >
>
> Thanks!
>
> Just a few comments:
>
> 1 =3D=3D=3D
>
> +       The authentication method used for authenticating the connection,=
 or
> +       NULL for background processes.
>
> What about? "NULL for background processes (except for parallel workers w=
hich
> inherit this information from their leader process)"

Ugh. That doesn't read very well at all to me. Maybe just "NULL for
background processes without a user"?


> 2 =3D=3D=3D
>
> +       cycle before they were assigned a database role.  Contains the sa=
me
> +       value as the identity part in <xref linkend=3D"system-user" />, o=
r NULL
> +       for background processes.
>
> Same comment about parallel workers.
>
> 3 =3D=3D=3D
>
> +# pg_stat_activity should contain trust and empty string for trust auth
> +$res =3D $node->safe_psql(
> +       'postgres',
> +       "SELECT auth_method, auth_identity=3D'' FROM pg_stat_activity WHE=
RE pid=3Dpg_backend_pid()",
> +       connstr =3D> "user=3Dscram_role");
> +is($res, 'trust|t', 'Users with trust authentication should show correct=
ly in pg_stat_activity');
> +
> +# pg_stat_activity should contain NULL for auth of background processes
> +# (test is a bit out of place here, but this is where the other pg_stat_=
activity.auth* tests are)
> +$res =3D $node->safe_psql(
> +       'postgres',
> +       "SELECT auth_method IS NULL, auth_identity IS NULL FROM pg_stat_a=
ctivity WHERE backend_type=3D'checkpointer'",
> +);
> +is($res, 't|t', 'Background processes should show NULL for auth in pg_st=
at_activity');
>
> What do you think about testing the parallel workers cases too? (I'm not =
100%
> sure it's worth the extra complexity though).

I'm leaning towards not worth that.

--=20
 Magnus Hagander
 Me: https://www.hagander.net/
 Work: https://www.redpill-linpro.com/