--- Begin Message ---
Subject: |
[PATCH 0/9] Removing 'make-forkexec-constructor/container' |
Date: |
Tue, 14 Nov 2023 15:05:51 +0100 |
Hello Guix!
This completes the removal of ‘make-forkexec-constructor/container’
as I intended to do with the introduction of ‘least-authority-wrapper’:
https://issues.guix.gnu.org/54997
The Jami use case exposed a few shortcomings, including those
addressed by <https://issues.guix.gnu.org/67172>, but thankfully
there are tests, which made it easier to validate the changes!
Feedback welcome!
Ludo’.
Ludovic Courtès (9):
services: pagekite: Use ‘least-authority-wrapper’.
services: pagekite: Add ‘configuration’ action.
services: bitlbee: Remove use of
‘make-forkexec-constructor/container’.
least-authority: Add support for changing UIDs/GIDs before exec.
tests: jami: Check status of Jami D-Bus session.
services: jami-dbus-session: Use ‘least-authority-wrapper’.
services: jami: Use ‘least-authority-wrapper’.
services: Remove unnecessary references to (gnu build shepherd).
shepherd: Remove ‘make-forkexec-constructor/container’.
gnu/build/shepherd.scm | 90 ----------------------
gnu/services/databases.scm | 41 +++++-----
gnu/services/messaging.scm | 77 ++++++++-----------
gnu/services/networking.scm | 36 +++++----
gnu/services/security-token.scm | 29 ++++---
gnu/services/telephony.scm | 132 +++++++++++++++++++-------------
gnu/services/web.scm | 50 ++++++------
gnu/tests/telephony.scm | 9 +++
guix/least-authority.scm | 25 +++++-
9 files changed, 223 insertions(+), 266 deletions(-)
base-commit: 08d94fe20eca47b69678b3eced8749dd02c700a4
prerequisite-patch-id: ea1da8834460072ad48cd7b4a3ec23e7205f2529
prerequisite-patch-id: eb3069189c1b61930a429f933fda673d8fe47691
--
2.41.0
--- End Message ---
--- Begin Message ---
Subject: |
Re: [bug#67175] [PATCH 7/9] services: jami: Use ‘least-authority-wrapper’. |
Date: |
Fri, 22 Dec 2023 00:42:19 +0100 |
User-agent: |
Gnus/5.13 (Gnus v5.13) |
Pushed as commit ca813173894360edef35a5d98878a3135e99e62a after
double-checking with:
make check-system TESTS="jami jami-provisioning jami-provisioning-partial"
I inserted a new commit, 2cc881ac13522566a27d996afd1fb88df363f75e, to
increase timeouts in the tests: on my laptop the three tests would
occasionally fail when using the initial 20s timeouts (it’s not “really”
20s as the time measured in the host, not in the guest).
Thanks,
Ludo’.
--- End Message ---