pg.ddx.io  pgsql-committers@postgresql.org mailing list archive  
help / color / mirror / Atom feed
From: Andres Freund <andres@anarazel.de>
To: Michael Paquier <michael@paquier.xyz>
Cc: pgsql-committers@lists.postgresql.org
Subject: Re: pgsql: Support "postgres -C" with runtime-computed GUCs
Date: Mon, 20 Sep 2021 20:20:40 -0700
Message-ID: <20210921032040.lyl4lcax37aedx2x@alap3.anarazel.de> (raw)
In-Reply-To: <E1mQglq-0007xB-Jw@gemulon.postgresql.org>
References: <E1mQglq-0007xB-Jw@gemulon.postgresql.org>

Hi,

On 2021-09-16 02:05:02 +0000, Michael Paquier wrote:
> Some TAP tests are added to provide some coverage here, using
> data_checksums in the tests of pg_checksums.

I just rebased my AIO branch ontop of this, and promptly got a CI
failure. Which appears to be unrelated to AIO. The CI system that we use
runs the scripts in a privileged account. Which normally is fine, because
pg_ctl drops permissions. However, the new test here doesn't.

See e.g. here https://cirrus-ci.com/task/6455395922214912?logs=bincheck#L27

# Running: postgres -D c:/cirrus/src/bin/pg_checksums/tmp_check/t_002_actions_node_checksum_data/pgdata -C data_checksums
not ok 22 - data_checksums=on is reported on an offline cluster status (got 1 vs expected 0)

#   Failed test 'data_checksums=on is reported on an offline cluster status (got 1 vs expected 0)'
#   at t/002_actions.pl line 189.
not ok 23 - data_checksums=on is reported on an offline cluster stdout /(?^:^on$)/

#   Failed test 'data_checksums=on is reported on an offline cluster stdout /(?^:^on$)/'
#   at t/002_actions.pl line 189.
#                   ''
#     doesn't match '(?^:^on$)'
not ok 24 - data_checksums=on is reported on an offline cluster stderr /(?^:database system is shut down)/

#   Failed test 'data_checksums=on is reported on an offline cluster stderr /(?^:database system is shut down)/'
#   at t/002_actions.pl line 189.
#                   'Execution of PostgreSQL by a user with administrative permissions is not
# permitted.
# The server must be started under an unprivileged user ID to prevent
# possible system security compromises.  See the documentation for
# more information on how to properly start the server.
# '
#     doesn't match '(?^:database system is shut down)'
### Starting node "node_checksum"


So the problem is that -C data_checksums requires permissions to be dropped,
but the way it's invoked here that's not guaranteed... And I don't think we
should remove that requirement, given that we're actually doing stuff to the
lock file...

Greetings,

Andres Freund





view thread (21+ messages)  latest in thread

Message-ID: <20210921032040.lyl4lcax37aedx2x@alap3.anarazel.de>
Permalink:  ../20210921032040.lyl4lcax37aedx2x@alap3.anarazel.de/
Also on:    postgresql.org/message-id/20210921032040.lyl4lcax37aedx2x@alap3.anarazel.de

 · 

reply

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Reply to all the recipients using the --to and --cc options:
  reply via email

  To: pgsql-committers@postgresql.org
  Cc: andres@anarazel.de, michael@paquier.xyz, pgsql-committers@lists.postgresql.org
  Subject: Re: pgsql: Support "postgres -C" with runtime-computed GUCs
  In-Reply-To: <20210921032040.lyl4lcax37aedx2x@alap3.anarazel.de>

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox