These tests have a tendency to fail with OOM on the autopkgtest
infrastructure. Increase QEMU_MEM to try and alleviate that.
Gbp-Pq: Name test-increase-QEMU_MEM-for-some-tests.patch
Origin: upstream, beff73f94e
Currently, the systemd-hwdb --root flag only has an effect for the
'update' verb. It would be useful to be able to use the --root option
for the 'query' verb too (e.g. for testing a hwdb.bin created with
systemd-hwdb update --root <path>).
Use sd_hwdb_new_from_path to initialize the hwdb if --root is passed to
systemd-hwdb query.
Note that this functionality was not added to 'udevadm hwdb' since that
command is deprecated.
Gbp-Pq: Name hwdb-implement-root-option-for-systemd-hwdb-query.patch
Origin: upstream, 60f0ba7556
The existing sd_hwdb_new function always initializes the hwdb from the
first successful hwdb.bin it finds from hwdb_bin_paths. This means there
is currently no way to initialize a hwdb from an explicit path, which
would be useful for systemd-hwdb query.
Add sd_hwdb_new_from_path to allow a sd_hwdb to be initialized from a
custom path outside of hwdb_bin_paths.
Gbp-Pq: Name sd-hwdb-add-sd_hwdb_new_from_path.patch
This test currently fails inside privileged LXD containers (LP: #1959013)
This reverts commit ae53f4b5e48860b473c4d05958486a77f84ecc6d.
Gbp-Pq: Name 0001-Revert-tests-add-test-case-for-UMask-BindPaths-combi.patch
blkid reports bcache superblock dev.uuid as a filesystem UUID but it actually
is not a filesystem, it's the UUID of the backing device, which is maintained
at /dev/bcache/by-uuid instead of /dev/disk/by-uuid.
[Forwarding Note]
There is an on-going discussion upstream whether this patch should exist.
This patch is not a FIX to LP: #1861941, but can work as a mitigation. The
FIX for LP: #1861941 is the bcache-tools (0003-Add-bcache-export-cached-
helper.patch).
Ryan Harper arguments are that blkid - and/or udev default rules - should skip
devices with "ID_FS_TYPE = bcache" by default from creating symlinks at
/dev/disk/{by-uuid,by-label}/{ID_FS_UUID_ENC,ID_FS_LABEL_ENC} just because
those devices aren't meant to be used directly (as they are backing devices
to bcache). Actually this is what was causing the issue fixed by bcache-tools
udev rules: symlink management for bcache backing devices were removing
/dev/bcache/xxx symlinks.
Considering that this is a minor delta, and I agree to Ryan's arguments, of
not having /dev/disk/by-uuid/xxx symlinks to devices that should not be
accessed directly, thus giving a better experience to end user, I'm keeping
this until either upstream provides it by default OR the patch
0003-Add-bcache-export-cached-helper.patch can be removed from bcache-tools
because udev and/or libblkid started differentiating UUID_CACHED and FS_UUID
when doing /dev/disk/ symlinks.
-
rafaeldtinoco
Author: Ryan Harper <ryan.harper@canonical.com>
Bug-Ubuntu: https://bugs.launchpad.net/bugs/1861941
Forwarded: https://github.com/systemd/systemd/pull/16317
Reviewed-by: Rafael David Tinoco <rafaeldtinoco@ubuntu.com>
Last-Update: 2020-07-23
Gbp-Pq: Name lp1861941-dont-generate-disk-byuuid-for-bcache-uuid.patch
The change broke networking in LXD containers
This reverts commit 96d7083c5499b264ecebd6a30a92e0e8fda14cd5.
LP: #1857123
Gbp-Pq: Name Revert-network-if-sys-is-rw-then-udev-should-be-around.patch
LP: #1895418
Thanks: Dimitri John Ledkov for the suggestion and
Co-authored-by: Dan Streetman <ddstreet@canonical.com>
Gbp-Pq: Topic debian
Gbp-Pq: Name UBUNTU-resolved-default-no-negative-caching.patch
This partially reverts commit 1beab8b0d0ff2d7d1436b52d4a0c3d56dc908962.
Until after the lowest LXD version running this packaged systemd contains
a6b7807033
The first LXD version fixed is 3.10. Ubuntu 18.04 LTS has LXD 3.0.3 and is
supported until 2028.
Gbp-Pq: Topic debian
Gbp-Pq: Name UBUNTU-Revert-namespace-be-more-careful-when-handling-namespacin.patch
Origin: other
Bug-Ubuntu: https://bugs.launchpad.net/snappy/+bug/1650688
Forwarded: not-needed (part of read-only /etc workaround)
Last-Update: 2021-09-24
Due to our read-only /etc workaround, the localtime link on such
system ends up in /etc/writable, not /etc. To make the link target
correct in both normal and such systems, makes the path absolute.
On Ubuntu Core, this eliminates the need for the wrapper script, and
makes the DBus interface work properly.
This patch can be dropped once LP: #1953172 got resolved
Last-Update: 2021-09-24
Gbp-Pq: Topic debian
Gbp-Pq: Name UBUNTU-Fix-timezone-setting-on-read-only-etc.patch
Origin: vendor, https://bugs.launchpad.net/snappy/+bug/1650688/comments/46
Bug-Ubuntu: https://bugs.launchpad.net/snappy/+bug/1650688
Forwarded: not-needed (part of read-only /etc workaround)
Last-Update: 2021-09-24
get_timezone() retrieve it by reading the link destination of
/etc/localtime, which on systems with read-only /etc will always point
to /etc/writable. Makes this function aware of the /etc/writable
redirection and handle it.
[ratchanan@ubports.com: add descrtiption and other metadata.]
This patch can be dropped once LP: #1953172 got resolved
Last-Update: 2021-09-24
Gbp-Pq: Topic debian
Gbp-Pq: Name timedatectl-lp1650688.patch
On Ubuntu Phone with readonly /etc we symlink
/etc/{adjtime,localtime,timezone,hostname,machine-info} to /etc/writable/, so
we need to update those files instead if the original files are symlinks into
/etc/writable/.
Forwarded: OMGno, this is a rather nasty hack until we fix system-image to get a writable /etc
Bug-Ubuntu: https://launchpad.net/bugs/1227520
Gbp-Pq: Topic debian
Gbp-Pq: Name UBUNTU-Support-system-image-read-only-etc.patch
On v4.4 kernels, on top of btrfs ephemeral lxd v3.0 containers generate this
other error code, instead of not supported. Skip the test for both error codes.
Gbp-Pq: Topic debian
Gbp-Pq: Name UBUNTU-test-sleep-skip-test_fiemap-upon-inapproriate-ioctl-.patch
lxd/lxc usually keep the usernamespace capabilities, whilst in practice one
does not have these in the initial namespace. Thus add additional condition
!container, such that sys-kernel-config.mount and systemd-modules.load.service
are not started in the lxd containers. This should make default lxd containers
start non-degraded.
Gbp-Pq: Topic debian
Gbp-Pq: Name UBUNTU-units-block-CAP_SYS_MODULE-units-in-containers-too.patch
When booting ppc64el virtual machines, they require seabios, unless -vga none
is specified. Since we do a direct kernel & initrd boot, with -nographic, we
really have no need for vga or seabios in this case.
Gbp-Pq: Topic debian
Gbp-Pq: Name UBUNTU-test-test-functions-launch-qemu-with-vga-none.patch
A change in apparmor mediates auto-activation attempts now through
AppArmor: https://cgit.freedesktop.org/dbus/dbus/commit/?id=dc25979eb
This breaks the snapd time{zone,server}-control interfaces which limt
sending dbus message to a (label=unconfined) org.freedesktop.timedate1
peers.
By adding the AssumedApparmorLabel=unconfined label the snapd interfaces
work again.
LP: #1749000
Gbp-Pq: Topic debian
Gbp-Pq: Name UBUNTU-Add-AssumedApparmorLabel-unconfined-to-timedate1-dbus.patch
As it fails to start in an unpriviledged container.
Gbp-Pq: Topic debian
Gbp-Pq: Name Ubuntu-units-set-ConditionVirtualization-private-users-on-j.patch
On Ubuntu, fallback DNS servers are disabled, therefore we do not leak queries
to a preset 3rd party by default. In resolved, dnssec is also disabled by
default, as too much of the internet is broken and using Ubuntu users to debug
the internet is not very productive - most of the time the end-user cannot fix
or know how to notify the site owners about the dnssec mistakes. Inherintally
the DHCP acquired DNS servers are therefore trusted, and are free to spoof
records. Not trusting DNS search domains, in such scenario, provides limited
security or privacy benefits. From user point of view, this also appears to be
a regression from previous Ubuntu releases which do trust DHCP acquired search
domains by default.
Therefore we are enabling UseDomains by default on Ubuntu.
Users may override this setting in the .network files by specifying
[DHCP|IPv6AcceptRA] UseDomains=no|route options.
Gbp-Pq: Topic debian
Gbp-Pq: Name Ubuntu-UseDomains-by-default.patch
If a package still ships only a SysV init script or if a service file or
tmpfile uses /var/run, downgrade those messages to debug. We can use
lintian to detect those issues.
For service files and tmpfiles in /etc, keep the warning, as those files
are typically added locally and aren't checked by lintian.
Closes: #981407
Gbp-Pq: Topic debian
Gbp-Pq: Name Downgrade-a-couple-of-warnings-to-debug.patch
When, for whatever reason, a scheduled shutdown fails to be set, systemd
will proceed with immediate shutdown without allowing the user to react.
This is counterintuitive because when a scheduled shutdown is issued,
it means the user wants to shutdown at a specified time in the future,
not immediately. This patch prevents the immediate shutdown and informs
the user that no action will be taken.
Fixes: #17575
Gbp-Pq: Topic debian
Gbp-Pq: Name systemctl-do-not-shutdown-immediately-on-scheduled-shutdo.patch
In Debian, late mounting of /usr is no longer supported, so it is safe
to install those files in /usr.
We want those facilities in /usr, not /, as this will make an eventual
switch to a merged-usr setup easier.
Closes: #971282
Gbp-Pq: Topic debian
Gbp-Pq: Name Move-sysusers.d-sysctl.d-binfmt.d-modules-load.d-back-to-.patch
We can't really control what helper programs are run from other udev
rules. E.g. running i386 binaries under amd64 is a valid use case and
should not trigger a SIGSYS failure.
Closes: #869719
Gbp-Pq: Topic debian
Gbp-Pq: Name udev-drop-SystemCallArchitectures-native-from-systemd-ude.patch
During package build, in minimal chroots, or other systems which do not already
have an /etc/machine-id we get six test failures. Introduce a
$SYSTEMD_MACHINE_ID_PATH environment variable which can specify a location
other than /etc/machine-id, so that the unit tests are independent from the
environment.
Also adjust test-fs-util to not assume that /etc/machine-id exists. Use
/etc/passwd instead which is created by base-files.
Closes: #851445
Bug: https://bugs.freedesktop.org/show_bug.cgi?id=62344
Gbp-Pq: Topic debian
Gbp-Pq: Name Add-env-variable-for-machine-ID-path.patch
Partially revert commit 15a900327ab as this completely breaks core dumps
without systemd-coredump. It's also contradicting core(8), and it's not
systemd's place to redefine the kernel definitions of core files.
Commit bdfd7b2c now honours the process' RLIMIT_CORE for systemd-coredump. This
isn't what RLIMIT_CORE is supposed to do (it limits the size of the core
*file*, but the kernel deliberately ignores it for piping), so set a static
2^63 core size limit for systemd-coredump to go back to the previous behaviour
(otherwise the change above would break systemd-coredump).
Bug-Debian: https://bugs.debian.org/815020
Gbp-Pq: Topic debian
Gbp-Pq: Name Revert-core-set-RLIMIT_CORE-to-unlimited-by-default.patch
Newer versions of initramfs-tools already fsck and mount / and /usr in
the initramfs. Skip the filesystem check in this case.
Based on a previous patch by Michael Biebl <biebl@debian.org>.
Closes: #782522Closes: #810748
Gbp-Pq: Topic debian
Gbp-Pq: Name Skip-filesystem-check-if-already-done-by-the-initram.patch
Global logic:
Add systemd-fsckd multiplexer which accepts multiple (via systemd-fsck's
/run/systemd/fsck.progress socket) fsck instances to connect to it and sends
progress report. systemd-fsckd then computes and writes to /dev/console the
number of devices currently being checked and the minimum fsck progress.
Plymouth and user interaction:
Forward the progress to plymouth and support canellation of in progress fsck.
Try to connect and send to plymouth (if running) some checked report progress,
using direct plymouth protocole.
Update message is the following:
fsckd:<num_devices>:<progress>:<string>
* num_devices corresponds to the current number of devices being checked (int)
* progress corresponds to the current minimum percentage of all devices being
checked (float, from 0 to 100)
* string is a translated message ready to be displayed by the plymouth theme
displaying the information above. It can be overridden by plymouth themes
supporting i18n.
Grab in fsckd plymouth watch key Control+C, and propagate this cancel request
to systemd-fsck which will terminate fsck.
Send a message to signal to user what key we are grabbing for fsck cancel.
Message is: fsckd-cancel-msg:<string>
Where string is a translated string ready to be displayed by the plymouth theme
indicating that Control+C can be used to cancel current checks. It can be
overridden (matching only fsckd-cancel-msg prefix) for themes supporting i18n.
Misc:
systemd-fsckd stops on idle when no fsck is connected.
Add man page explaining the plymouth theme protocol, usage of the daemon
as well as the socket activation part. Adapt existing fsck man page.
Note that fsckd had lived in the upstream tree for a while, but was removed.
More information at
http://lists.freedesktop.org/archives/systemd-devel/2015-April/030175.html
-
Gbp-Pq: Topic debian
Gbp-Pq: Name fsckd-daemon-for-inter-fsckd-communication.patch
logind fails to start in environments without dbus, such as LXC containers or
servers. Add a startup condition to avoid the very noisy startup failure.
Consider both dbus-daemon (the reference implementation) and
dbus-broker.
Part of #772700
Gbp-Pq: Topic debian
Gbp-Pq: Name Only-start-logind-if-dbus-is-installed.patch
Revert upstream commit 46b131574fdd7d77 for now, until Debian's sysloggers
can/do all read from the journal directly. See
http://lists.freedesktop.org/archives/systemd-devel/2014-November/025550.html
for details. Once we grow a journal.conf.d/ directory, sysloggers can be moved
to pulling from the journal one by one and disable forwarding again in such a
conf.d snippet.
Gbp-Pq: Topic debian
Gbp-Pq: Name Re-enable-journal-forwarding-to-syslog.patch
The /run/lock directory is world-writable in Debian due to historic
reasons. To avoid user processes filling up /run, we mount a separate
tmpfs for /run/lock. As this directory needs to be available during
early boot, we make it an API fs.
Drop it from tmpfiles.d/legacy.conf to not clobber the permissions.
Closes: #751392
Gbp-Pq: Topic debian
Gbp-Pq: Name Make-run-lock-tmpfs-an-API-fs.patch
Use /etc/default/locale instead of /etc/locale.conf for locale settings.
Use /etc/default/keyboard instead of /etc/X11/xorg.conf.d/00-keyboard.conf for
keyboard configuration.
Read/write /etc/timezone if /etc/localtime does not exist.
Gbp-Pq: Topic debian
Gbp-Pq: Name Use-Debian-specific-config-files.patch
Introduce rootpkglibdir for installing libsystemd-{shared,core}.so.
The benefit over using rootlibexecdir is that this path can be
multiarch aware, i.e. this path can be architecture qualified.
This is something we'd like to make use of in Debian/Ubuntu to make
libsystemd-shared co-installable, e.g. for i386 the path would be
/usr/lib/i386-linux-gnu/systemd/libsystemd-shared-*.so and for amd64
/usr/lib/x86_64-linux-gnu/systemd/libsystemd-shared-*.so.
This will allow for example to install and run systemd-boot/i386 on an
amd64 host. It also simplifies/enables cross-building/bootstrapping.
For more infos about Multi-Arch see https://wiki.debian.org/Multiarch.
See also https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=990547
(cherry picked from commit 5fb225615bf751b97644bed7aae44f69ba03cc84)
Gbp-Pq: Name meson-install-libsystemd-shared-into-rootpkglibdir.patch
Those binaries aren't needed during early boot.
(cherry picked from commit 003a67616148a8c2b94aa0c87595465f5dcac508)
Gbp-Pq: Name Move-homectl-and-userdbctl-to-bindir.patch
This fixes#20820
(cherry picked from commit 87c7779599ade84df7934c563adb7fe550c871f7)
Gbp-Pq: Name Do-not-require-a-valid-version-when-parsing-sd-boot-loade.patch
Nick Rosbrook