Daan De Meyer [Thu, 5 Sep 2024 16:38:33 +0000 (18:38 +0200)]
mkosi: Install util-linux-script on Rawhide
It's now subpackaged so we can build images without pulling in
libutempter but we use script in the testsuite so let's install the
subpackage.
(cherry picked from commit
4eae2be0d7cd8832688d38ef2cdf4a42578fbff0)
Daan De Meyer [Sat, 7 Sep 2024 10:13:58 +0000 (12:13 +0200)]
mkosi: Drop _fixperms workaround
This was added to deal with a bug in the rpm 4.20 rc in Rawhide
but since that's been fixed, let's drop the workaround.
(cherry picked from commit
517e8922101557fccf4a7f8fc922e5102e787ec7)
Yu Watanabe [Fri, 6 Sep 2024 06:00:32 +0000 (15:00 +0900)]
tree-wide: check if non-empty password is acquired
(cherry picked from commit
204529d0fcde9a95119b489225620f36649c2f5b)
Yu Watanabe [Fri, 6 Sep 2024 02:19:39 +0000 (11:19 +0900)]
ask-password: refuse empty password strv
Fixes #34270.
(cherry picked from commit
623a8b1922bdbd2fb06bc5f2c67d3d6930efc58e)
Yu Watanabe [Thu, 5 Sep 2024 08:57:24 +0000 (17:57 +0900)]
test: fix indentation
(cherry picked from commit
fe6049d0210c89a595ae598d87dcefe7bfbe3a1d)
Yu Watanabe [Thu, 5 Sep 2024 08:58:29 +0000 (17:58 +0900)]
test: add test case for systemd-repart --seed=random
For issue #34257.
(cherry picked from commit
56d6ebd40468e2a743b39ad7d87b0675bdf9a042)
Yu Watanabe [Thu, 5 Sep 2024 06:49:09 +0000 (15:49 +0900)]
repart: initialize seed earlier
As the seed is used by context_load_partition_table() -> derive_uuid().
Fixes #34257.
(cherry picked from commit
b8a8000aba4620a1d93ac1d04dbb86050d9c5fe8)
Yu Watanabe [Thu, 5 Sep 2024 06:05:32 +0000 (15:05 +0900)]
nspawn: refuse to bind mount device node from host when --private-users= is specified
Also do not chown if a device node is bind-mounted.
Fixes #34243.
(cherry picked from commit
efedb6b0f3cff37950112fd37cb750c16d599bc7)
Mike Yuan [Sat, 31 Aug 2024 13:42:43 +0000 (15:42 +0200)]
audit-util: check correct errno
(cherry picked from commit
190a0953808608b099f9465f9e786e4efe276c26)
Daan De Meyer [Thu, 29 Aug 2024 20:59:48 +0000 (22:59 +0200)]
repart: Keep existing directory timestamps intact when copying
Otherwise, when merging multiple directory trees, the output becomes
unreproducible as the directory timestamps will be changed to the current
time when copying identical directories from the second tree.
We introduce a new copy flag to achieve this behavior.
(cherry picked from commit
d850a544bc1f895decb452160c97a884a20b12b7)
Daan De Meyer [Wed, 28 Aug 2024 12:08:30 +0000 (14:08 +0200)]
ukify: Skip test on architectures without UEFI
(cherry picked from commit
5121f7c45b37afca53c89f42123b1dd6a04fa80f)
Ronan Pigott [Mon, 26 Aug 2024 08:39:20 +0000 (01:39 -0700)]
resolved: clear the AD bit for bypass packets
When the bypass logic is invoked, such as for queries to the stub with
the DO bit set, be certain to clear the AD bit in the reply before
forwarding it if the answer is not known to be authentic.
(cherry picked from commit
13e15dae9f0b4566d3ea2ed058a5dd44751216da)
Yu Watanabe [Fri, 23 Aug 2024 01:14:45 +0000 (10:14 +0900)]
udevadm/test,test-builtin: enable debugging logs by default again
The lines were mistakenly dropped by
aa976d87889ae22b7347787a1ebd03a31dcc0a9e.
(cherry picked from commit
b9142e2ba7fce5455fd3a782c69a45a0d8eaf866)
Mike Yuan [Sat, 17 Aug 2024 16:34:46 +0000 (18:34 +0200)]
logind-session: downgrade user@.service dep to Wants=
This partially reverts
52bcc872b5f97a14a9b4e4e383f45bb3066e1643.
We explicitly support running without user manager,
hence only user-runtime-dir@.service should be
required.
Fixes #33405
(cherry picked from commit
26f78eff6905179838f311e4f4e2809677c0bb78)
Alyssa Ross [Wed, 21 Aug 2024 12:21:47 +0000 (14:21 +0200)]
bootctl: don't load etc/machine-info from cwd
arg_root defaults to null, so if --root isn't given, this would try reading
etc/machine-info from the current working directory, which is likely to fail.
Fixes:
77db9ef2ab ("boot: Make sure we take --root into account everywhere.")
(cherry picked from commit
0452779b0054f5c2724b745b1db33bba1ac8e677)
maia x. [Mon, 19 Aug 2024 19:47:21 +0000 (12:47 -0700)]
namespace: Fix extension release memory leak
In apply_one_mount(), in the MOUNT_EXTENSION_DIRECTORY case,
char **extension_release was used as a return pointer twice but only
cleaned up once in the end. Fix it by removing duplicate code that
was causing this issue.
Fixes issue introduced in
55ea4ef096543d2bceea9315868d5aca945d7a57.
(cherry picked from commit
010ea061fceb84d36259d576f52c09b940d0d615)
Kornilios Kourtis [Thu, 15 Aug 2024 15:22:35 +0000 (17:22 +0200)]
process-util: handle pidfd_spawn() returning E2BIG
In some kernels (specifically, 5.4) even though the clone3 syscall is
supported, setting CLONE_INTO_CGROUP is not. The error message returned
in this case is E2BIG.
If posix_spawn_wrapper encounters this error, it does not retry, and
cannot spawn any programs in said kernels.
This commit adds a check for the E2BIG error and retries pidfd_spawn()
without the POSIX_SPAWN_SETCGROUP flag.
If we encounter an E2BIG error, and the pidfd_spawn() succeeds after
removing the POSIX_SPAWN_SETCGROUP flag, then we cache the result so
that we do not retry every time.
Originally, this issue was reported in https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=
1077204.
Signed-off-by: Kornilios Kourtis <kornilios@gmail.com>
(cherry picked from commit
7ac58157ca67ab001307f1fd72e0cc7c0c4e846a)
Yu Watanabe [Wed, 7 Aug 2024 08:48:42 +0000 (17:48 +0900)]
test: add ASSERT_OK_POSITIVE() and ASSERT_OK_ZERO()
(cherry picked from commit
538766ddf47cfd457d9c69f108c035d4d97eaba1)
Raphaël Mélotte [Fri, 14 Jun 2024 12:37:29 +0000 (14:37 +0200)]
src/basic/missing_loop.h: fix missing LOOP_SET_BLOCK_SIZE
Builds with kernels headers < 4.14 fail with:
../src/shared/loop-util.c: In function ‘loop_configure_fallback’:
../src/shared/loop-util.c:237:31: error: ‘LOOP_SET_BLOCK_SIZE’ undeclared (first use in this function); did you mean ‘LOOP_SET_DIRECT_IO’?
if (ioctl(fd, LOOP_SET_BLOCK_SIZE, (unsigned long) c->block_size) < 0)
^~~~~~~~~~~~~~~~~~~
LOOP_SET_DIRECT_IO
Fixes: https://github.com/systemd/systemd/issues/33341
Signed-off-by: Raphaël Mélotte <raphael.melotte@mind.be>
(cherry picked from commit
56ab1c54497d9fac74380ff9e11aaf931a917d2b)
Yu Watanabe [Fri, 16 Aug 2024 16:48:50 +0000 (01:48 +0900)]
test: add test case that 'nspawn --network-veth' enables IP forwarding
(cherry picked from commit
08779d7c55add3a3bad5228d35012c15e92c47a5)
Yu Watanabe [Fri, 16 Aug 2024 16:47:33 +0000 (01:47 +0900)]
test: allow to skip matrix_run_one() if $TEST_MATCH_TESTCASE is set
(cherry picked from commit
7908e1d459f5f2893d6aaf1d62009da7856f9410)
Yu Watanabe [Fri, 16 Aug 2024 15:00:32 +0000 (00:00 +0900)]
network: make IPMasquerade= imply global IP forwarding settings again
After
3976c430927e1bfefa0413f80ebac84ab9a64350 (#31423), IPMasquerade=
implies only per-interface IP forwarding. That means, nspawn users need
to manually enable IPv4/IPv6Forwarding= in networkd.conf when
--network-veth or friend is used. Even the change was announced in NEWS,
the change itself breaks backward compatibility and extremely reduces
usability.
Let's make the setting imply the global setting again.
Fixes #34010.
(cherry picked from commit
0b695febb22ea5701eab4aee801e8a861ffdbaa6)
Yu Watanabe [Fri, 16 Aug 2024 14:30:02 +0000 (23:30 +0900)]
network/lldp-tx: introduce link_lldp_tx_update_capabilities()
Currently it is unused, but it will be used later.
(cherry picked from commit
8ceca83141ffea2e383d1205957bdc47f5492875)
Yu Watanabe [Fri, 16 Aug 2024 14:29:36 +0000 (23:29 +0900)]
sd-lldp-tx: insert missing empty line
(cherry picked from commit
fbcd7e054bd4bc3e25afa41535d08ab2f2f0a095)
Daan De Meyer [Mon, 19 Aug 2024 08:53:53 +0000 (10:53 +0200)]
Merge pull request #34032 from DaanDeMeyer/backport-tests
Backport various commits to v256-stable
Daan De Meyer [Wed, 14 Aug 2024 10:43:05 +0000 (12:43 +0200)]
Add $SYSTEMD_IN_CHROOT to override chroot detection
When running unprivileged, checking /proc/1/root doesn't work because
it requires privileges. Instead, let's add an environment variable so
the process that chroot's can tell (systemd) subprocesses whether
they're running in a chroot or not.
(cherry picked from commit
2701c2f67de592fe2565b084cfaec1667767dbaf)
Daan De Meyer [Sun, 18 Aug 2024 22:13:47 +0000 (00:13 +0200)]
test-dhcp-server: Gracefully handle the network being down
(cherry picked from commit
4cf7a676af9a79ff418227d8ff488dfca6f243ab)
Daan De Meyer [Sun, 18 Aug 2024 11:20:14 +0000 (13:20 +0200)]
test: Gracefully handle running within user namespace with single user
Unprivileged users often make themselves root by unsharing a user namespace
and then mapping their current user to root which does not require privileges.
Let's make sure our tests don't fail in such an environment by adding checks
where required to see if we're not running in a user namespace with only a
single user.
(cherry picked from commit
ef31767ed7e21672a50b77e7b3935948aaba114c)
Daan De Meyer [Sun, 18 Aug 2024 11:19:30 +0000 (13:19 +0200)]
test-netlink: Gracefully handle the loopback interface being down
(cherry picked from commit
d098b8df6e5c1b4c834272dd1397345483116db6)
Luca Boccassi [Thu, 15 Aug 2024 20:46:49 +0000 (21:46 +0100)]
meson: bump version to 256.5
Daan De Meyer [Thu, 15 Aug 2024 20:22:56 +0000 (22:22 +0200)]
Merge pull request #34003 from bluca/v256-stable
v256 batch
Luca Boccassi [Wed, 14 Aug 2024 21:49:10 +0000 (22:49 +0100)]
test: add TEST_SKIP to mkosi integration test wrapper
Takes a space-separate list of test names, allows to skip one or more tests
(cherry picked from commit
ddc91af4eaa32511f92c83b2c24d9cc0425fd5f5)
Yu Watanabe [Thu, 15 Aug 2024 07:14:37 +0000 (16:14 +0900)]
test: remove temporary directory for test-ukify on success
(cherry picked from commit
e9a5b4a10eaa10fd43c69d148d57c7f4d8e10a4f)
Yu Watanabe [Thu, 15 Aug 2024 06:58:29 +0000 (15:58 +0900)]
test: remove temporary directory on success
Also use mkdtemp_open() with specific template.
Follow-up for
e7e52ff9b6d6bbfcdcc298ef3c156420b51d58b8.
(cherry picked from commit
63be834db8fce69536b3ec04f636f1b838dd531a)
Daan De Meyer [Wed, 14 Aug 2024 14:51:11 +0000 (16:51 +0200)]
mkosi: Fix debian/not-installed build logic
- Expand globs
- Filter out directories
- Append to the list of packaged files, not the list of installed files
Fixes the issue in https://github.com/systemd/systemd/pull/32363
(cherry picked from commit
0409213cbc1f939b6f2f72deabc4c92a7fe30523)
Chengen Du [Mon, 12 Aug 2024 03:41:52 +0000 (11:41 +0800)]
udev: Handle PTP device symlink properly on udev action 'change'
PTP device symlink creation rules are currently executed only when the
udev action is 'add'. If a user reloads the rules and runs the udevadm
trigger command to reapply changes, the symlink may be deleted, which
can prevent the chronyd service from restarting properly.
Signed-off-by: Chengen Du <chengen.du@canonical.com>
(cherry picked from commit
6bd12be3fa7761f190e17efdbdbff4440da7528b)
Thorsten Scherer [Wed, 14 Aug 2024 20:32:44 +0000 (22:32 +0200)]
repart: Fix misleading typo in GPT partition flag
Bit 60 is the one corresponding to ReadOnly, not 50. Fix this.
(cherry picked from commit
932cc94436e653d0487c29e0dd44685610cd7bcb)
Daan De Meyer [Thu, 8 Aug 2024 20:34:05 +0000 (22:34 +0200)]
test: Use usual setup in integration-test-setup script
(cherry picked from commit
db043a9b1a109478581f19b69a3019fc15f672aa)
Daan De Meyer [Thu, 8 Aug 2024 19:49:09 +0000 (21:49 +0200)]
test: Fix section of StateDirectory=
(cherry picked from commit
dba138b863fadf57362026281d0597b2d7736090)
Daan De Meyer [Thu, 8 Aug 2024 19:48:34 +0000 (21:48 +0200)]
mkosi: Install llvm
For llvm-symbolizer which is required to sanitize address sanitizer
reports.
(cherry picked from commit
c022fcd80c5bf94e8ac5511559f085303b24b070)
Daan De Meyer [Thu, 8 Aug 2024 08:59:08 +0000 (10:59 +0200)]
mkosi: Disable pagination in gdb
(cherry picked from commit
7628565604f5a6a572cb4a33ccde9a64fcc9ff09)
James Muir [Tue, 13 Aug 2024 01:53:24 +0000 (21:53 -0400)]
man/systemd-sysusers, man/systemd-tmpfiles: fix sentence fragments
add "are executed" so that those two sentences have a verb.
(cherry picked from commit
c5903cde63bb21a37c3e841c160d1fa060ccf7f2)
Daan De Meyer [Tue, 13 Aug 2024 08:36:40 +0000 (10:36 +0200)]
docs: Mention the new mount API in the container interface doc
Let's mention that the new mount API may be used to establish new
mounts in a container without needing the /run/host/incoming directory.
(cherry picked from commit
74cc5e2041a2c32e1824b32316bd95f2c8a811f5)
Mike Yuan [Sun, 11 Aug 2024 13:41:30 +0000 (15:41 +0200)]
edit-util: do not try to recreate temp file if missing
We initially read from temp file, then strip it, and write
back to it. If the file suddenly disappeared during the process,
it indicates someone else is touching our temp file
behind our back. Let's not silently continue.
(cherry picked from commit
3b5b2ff8fa6413afc2e5a058ea8281a58a4c691d)
Ivan Shapovalov [Wed, 7 Aug 2024 08:02:45 +0000 (10:02 +0200)]
core/exec-invoke: call setpriority() after sched_setattr()
The nice value is part of struct sched_attr, and consequently invoking
sched_setattr() after setpriority() would clobber the nice value with
the default (as we are not setting it in struct sched_attr).
It would be best to combine both calls, but for now simply invoke
setpriority() after sched_setattr() to make sure Nice= remains effective
when used together with CPUSchedulingPolicy=.
(cherry picked from commit
711a157738b3dcd29a5ebc8f498eb46bfac59652)
Nick Rosbrook [Wed, 7 Aug 2024 22:18:06 +0000 (18:18 -0400)]
core/unit: do not use unit path cache in unit_need_daemon_reload()
When unit_need_daemon_reload() calls unit_find_dropin_paths() to check
for new drop-in configs, the manager's unit path cache is used to limit
which directories are considered. If a new drop-in directory is created,
it may not be in the unit path cache, and hence unit_need_daemon_reload()
may return false, despite a new drop-in being present. However, if a
unit path cache is not given to unit_file_find_dropin_paths() at all,
then it behaves as if the target path was found in the unit path cache.
So, to fix this, adapt unit_find_dropin_paths() to take a boolean
argument indicating whether or not to pass along the unit path cache.
Set this to false in unit_need_daemon_reload().
Fixes #31752
(cherry picked from commit
82c482d573c9d2f3ab36f7be8d32772f90f2c335)
Luca Boccassi [Thu, 8 Aug 2024 14:13:38 +0000 (15:13 +0100)]
test: note in README how to get full list of Ubuntu CI jobs
(cherry picked from commit
79488ac0038aca5fca5b2938eb645d35015664cf)
Daan De Meyer [Wed, 7 Aug 2024 19:23:37 +0000 (21:23 +0200)]
mkosi: Disable debuginfod
We generally don't care about library debuginfo so let's just disable
debuginfod so it doesn't get in the way when debugging.
We use /root/.gdbinit as the systemwide gdbinit location is distribution
specific.
(cherry picked from commit
2561e2a35601383bfba30da58d378303cb9e39aa)
Yu Watanabe [Wed, 7 Aug 2024 05:03:13 +0000 (14:03 +0900)]
meson: enable -Wunterminated-string-initialization
With
af1a6db58fde8f64edcf7d27e1f3b636c999934c, now we can build with the
option.
(cherry picked from commit
f548bc4011bcdab008b125b9d0993817efa00718)
Cristian Rodríguez [Sun, 4 Aug 2024 22:51:54 +0000 (18:51 -0400)]
basic|boot: silence Wunterminated-string-initialization gcc15 warnings
gcc15 has -Wunterminated-string-initialization in -Wextra and
warns about string constants that are not null terminated even though
the functions do do out of bounds access.
Silence the warnings by simply not providing an explicit size.
(cherry picked from commit
af1a6db58fde8f64edcf7d27e1f3b636c999934c)
Nick Rosbrook [Tue, 6 Aug 2024 00:43:15 +0000 (20:43 -0400)]
sysusers: check if requested group name matches user name in queue
When creating a user, check if the requested group name matches a user
name in the queue. If that matched user name is also going to be a group
name, then use it for the new user too. In other words, allow the
following:
u foo -
u bar -:foo
when both foo and bar are new users.
Fixes #33547
(cherry picked from commit
18a8f03e5160ca3828d327d9bbd1b32f26d792a3)
Yu Watanabe [Sun, 4 Aug 2024 05:09:53 +0000 (14:09 +0900)]
systemctl: gracefully adjust bus transport and runtime scope when --boot-loader-entry=help
This fixes the following assertion:
===
SYSTEMD_LOG_LEVEL=debug systemctl --user -H foo --boot-loader-entry=help
Assertion 'transport != BUS_TRANSPORT_REMOTE || runtime_scope == RUNTIME_SCOPE_SYSTEM' failed at src/shared/bus-util.c:284, function bus_connect_transport(). Ignoring.
Failed to connect to bus: Operation not supported
===
Fixes a bug introduced by
97af80c5a7029c3f92e982dcf9338b9e67ad9cde.
Fixes #33661.
Fixes oss-fuzz#70153.
(cherry picked from commit
1056457d1132e78cbf914e8abfa661465b952412)
Arnaud Patard [Mon, 8 Jul 2024 13:39:14 +0000 (15:39 +0200)]
src/pcrlock/pcrlock.c: Handle empty pcrlock.d directories
Running the following commands:
# mkdir -p /var/lib/pcrlock.d/123-empty.pcrlock.d
# /usr/lib/systemd/systemd-pcrlock predict --pcr=1+2+3+4+5+16
Will result in:
...
Floating point exception
Running the following commands:
# mkdir -p /var/lib/pcrlock.d/123-empty.pcrlock.d
# /usr/lib/systemd/systemd-pcrlock make-policy --pcr=1+2+3+4+5+16
Will result to this (partial) log:
...
Predicted future PCRs in 133us.
[]
...
Written policy digest
0000000000000000000000000000000000000000000000000000000000000000 to NV index 0x1921da6
...
So, add missing checks to handle gracefully cases where there's no variant
inside the component.
Signed-off-by: Arnaud Patard <arnaud.patard@collabora.com>
(cherry picked from commit
e7a93e75219b22424bab95fe45982f5eef21d581)
Daan De Meyer [Mon, 5 Aug 2024 13:37:46 +0000 (15:37 +0200)]
mkosi: Prevent busybox from getting pulled into opensuse images
OpenSUSE's busybox has a bunch of Provides for basic tools that cause
it to get pulled into images unless the corresponding tool is explicitly
installed so let's add explicit tools to make sure we don't get busybox.
(cherry picked from commit
857e4528f5fe556bbe98e32ffcde8932d0d80f83)
Daan De Meyer [Fri, 2 Aug 2024 14:25:03 +0000 (16:25 +0200)]
test: Add a way to quickly iterate on an integration test
Rebuilding the integration test every time is very slow. Let's
introduce a way to iterate on an integration test without rebuilding
the image every time. By making a btrfs snapshot before we run the
integration test, we can then systemctl soft-reboot after running
the test to restore the rootfs to a pristine state before running
the test again.
As /run/nextroot will get nuked on reboot or soft-reboot, we introduce
a tmpfiles snippet to make sure it is recreated every (soft-)reboot
and adapt the existing tests to deal with this new symlink.
(cherry picked from commit
af153e36ae67c242251951c12d6d6b6ae4783845)
Daan De Meyer [Fri, 2 Aug 2024 14:23:40 +0000 (16:23 +0200)]
mkosi: Switch back to btrfs
The next commit will introduce a way to iterate on integration
tests which depends on btrfs specific features.
We leave CentOS on ext4 as its kernel does not support btrfs.
(cherry picked from commit
edc6592e532806fe9ac28400138d5d4e1492a7d0)
Daan De Meyer [Fri, 2 Aug 2024 14:22:37 +0000 (16:22 +0200)]
mkosi: Enable Autologin= again on Debian
Debian uses /usr/bin/login from the shadow package instead of util-linux
which doesn't support credentials. Let's enable autologin the old
fashioned way for now.
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=833256
(cherry picked from commit
ea696c5f5300c240ae2749f7791aaa78837d556b)
Daan De Meyer [Fri, 2 Aug 2024 13:46:41 +0000 (15:46 +0200)]
test: Implement TEST_SHELL for mkosi based test runner
TEST_SHELL can be used to get a shell in the integration test
environment without actually immediately starting the test.
(cherry picked from commit
dd1c01b20fb8eef6a4d41ac7f0d3d83399e27593)
Daan De Meyer [Fri, 2 Aug 2024 14:16:14 +0000 (16:16 +0200)]
test: Rename INTERACTIVE_DEBUG to TEST_SHELL
(cherry picked from commit
33f400a9e05f3a09d28ac3d8251cf82592a207c6)
Daan De Meyer [Fri, 2 Aug 2024 13:18:45 +0000 (15:18 +0200)]
docs: Update upgrade commands in HACKING.md
- Add the required options to make the package managers non interactive
- Use apt-get instead of apt
- Remove --reinstall from apt-get command so we only install newer packages
- Add --needed to pacman command so we only install newer packages
(cherry picked from commit
5f5b6fa9017c7c5de1e8a3e6b8888fd71a1d3862)
Daan De Meyer [Fri, 2 Aug 2024 11:45:41 +0000 (13:45 +0200)]
mkosi: Drop locale tmpfiles snippet
Not required anymore as we don't do /usr only anymore and the symlink
will be created by the debian systemd package.
(cherry picked from commit
63bafef12ba867010507963d5aac67f4d64d85e9)
Yu Watanabe [Mon, 5 Aug 2024 05:18:21 +0000 (14:18 +0900)]
journal: set flushed flag even if we fail to open runtime journals
As at this stage, a persistent journal file has been already opened, and
saved seqnum has been reset, and any later journal entries will be stored
to the file. Hence we should not open the runtime journal file by
server_system_journal_open() again.
(cherry picked from commit
b55027efe41b266df4c4704e924fff7dfb32ad70)
Yu Watanabe [Mon, 5 Aug 2024 06:24:07 +0000 (15:24 +0900)]
journal: comment the default value in journald.conf
(cherry picked from commit
0d113f8e70243c1a8f0587105195e51e027a4725)
Yu Watanabe [Mon, 5 Aug 2024 05:56:11 +0000 (14:56 +0900)]
journal: do not rotate journal when MaxRetentionSec= is set
The setting is about vacuuming archived journal files. It is not
necessary to rotate the current journal. Note, journal file rotation is
controlled by MaxFileSec=.
Fixes #31315.
(cherry picked from commit
b63c18db03d0c120ba9eaa2aa9d0d43d80eea02b)
Yu Watanabe [Mon, 5 Aug 2024 01:46:41 +0000 (10:46 +0900)]
resolve: refuse invalid service without type field
Fixes Fixes #33935.
(cherry picked from commit
b48ab08732a76b7337628e1e716f11c687000903)
Yu Watanabe [Sun, 4 Aug 2024 05:07:13 +0000 (14:07 +0900)]
systemctl: refuse --capsule=foo with --system
Fixes the following assertion:
===
systemctl --capsule=hoge --system reboot
Assertion 'runtime_scope == RUNTIME_SCOPE_USER' failed at src/shared/bus-util.c:479, function bus_connect_transport(). Aborting.
Aborted (core dumped)
===
Follow-up for
56cb74c3cd1358d7d0b3f613feaf2eeab601a6bd.
(cherry picked from commit
bcf982223cc5d051818dc88866a42f6a08f39a4a)
Luca Boccassi [Sat, 3 Aug 2024 23:00:05 +0000 (00:00 +0100)]
base-filesystem: do not attempt to create a /lib64 -> /usr/lib/<tuple> symlink
In multi-arch distributions (debian and derivatives) multiarch tuples under
/usr/lib are used, such as /usr/lib/x86_64-linux-gnu/ but the /lib64 symlink
should never point there, it should always point to /usr/lib64, as that's
how they are set up by distribution-specific tools.
https://packages.debian.org/bookworm/amd64/libc6-i386/filelist
https://packages.debian.org/bookworm/mipsel/libc6-mips64/filelist
https://salsa.debian.org/md/usrmerge/-/blob/master/convert-usrmerge?ref_type=heads#L295
https://salsa.debian.org/md/usrmerge/-/blob/master/convert-usrmerge?ref_type=heads#L517
http://bugs.debian.org/
1076491
Fixes https://github.com/systemd/systemd/issues/33919
(cherry picked from commit
b75c13731ee0867a8d7889348fc8da1869af7551)
Daan De Meyer [Sat, 3 Aug 2024 18:10:54 +0000 (20:10 +0200)]
meson: Use -fstrict-flex-arrays=3
Let's explicitly pass the value to -fstrict-flex-arrays. This does
not change behavior but it does (selfishly) make my error not bug
out with an error saying -fstrict-flex-arrays does not exist.
(cherry picked from commit
ad723ca3e5bd41d2d884760375534910bb55d9b3)
Mike Yuan [Sat, 3 Aug 2024 20:38:18 +0000 (22:38 +0200)]
core/execute-serialize: use serialize_item_escaped() for external paths
Otherwise, read_stripped_line() would spuriously drop trailing spaces.
Fixes #33924
(cherry picked from commit
9be46b1da8b01c3f47e6c050185f2b45484d6300)
Mike Yuan [Sat, 3 Aug 2024 20:37:41 +0000 (22:37 +0200)]
core/execute-serialize: drop extraneous '=' in ip-{in,e}gress serialization
(cherry picked from commit
f0fdd13c2f06f9c78747103b971566e2c62b9333)
Yu Watanabe [Fri, 2 Aug 2024 07:45:15 +0000 (16:45 +0900)]
man/net-naming-scheme: mention that NAMING_BRIDGE_MULTIFUNCTION_SLOT is reverted
Follow-up for
af7417ac7b07bc01232982bf46e9d72e69e7f820.
Closes #33596.
(cherry picked from commit
1c0130e8dc3c99d5a85be41e9172adb0ff0cf7fd)
Yu Watanabe [Fri, 2 Aug 2024 04:42:29 +0000 (13:42 +0900)]
man: extend explanation for ConfigureWithoutCarrier= in systemd.network(5)
Prompted by #33702.
(cherry picked from commit
347c8822d1a8a5b70624920b3de2a91d4e0fca91)
Daniel P. Berrangé [Fri, 2 Aug 2024 12:17:56 +0000 (13:17 +0100)]
man/systemd-detect-virt: list known CVM technologies
Add a section which lists the known confidential virtual machine
technologies.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
(cherry picked from commit
a8fb5d21fd6127a6d05757c793cc9ba47f65c893)
Daniel P. Berrangé [Fri, 2 Aug 2024 10:03:10 +0000 (11:03 +0100)]
confidential-virt: add detection for s390x target
The s390x platform provides confidential VMs using the "Secure Execution"
technology, which is also referred to as "Protected Virtualization" or
just "prot virt" in Linux / QEMU.
This can be detected through a simple sysfs attribute.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
(cherry picked from commit
6c35e0a51cc6a852ce239ea46cd75c133212a68e)
Daniel P. Berrangé [Fri, 2 Aug 2024 15:26:00 +0000 (16:26 +0100)]
confidential-virt: split caching of CVM detection into separate method
We have different impls of detect_confidential_virtualization per
architecture. The detection is cached in the x86_64 impl, and as we
add support for more targets, we want to use caching for all. It thus
makes sense to split caching out into an architecture independent
method.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
(cherry picked from commit
1c4bd7adcc281af2a2dd40867f64f2ac54a43c7a)
Frantisek Sumsal [Fri, 2 Aug 2024 08:13:47 +0000 (10:13 +0200)]
test: don't use /skipped for subtests
Since, at least the old framework, checks for the presence of the file
at the end and marks the whole test as skipped if it exists.
Resolves: systemd/systemd-centos-ci#728
(cherry picked from commit
4d1fbe53c1cc9b04acd897c728d996191daa618d)
Ronan Pigott [Thu, 1 Aug 2024 17:59:12 +0000 (10:59 -0700)]
resolved: don't treat conn reset as packet loss
tcp reset / icmp port-unreachable are markedly different conditions than
packet loss. It doesn't make much sense to retry in this case. It's
actually not clear if there is any benefit at all retrying tcp
connections, which were presumably already retried as necessary by the
tcp stack.
(cherry picked from commit
ddd710a355acc698b48159f3e501dda5a7dc2704)
Yu Watanabe [Thu, 1 Aug 2024 03:03:54 +0000 (12:03 +0900)]
import: check overflow
Fixes CID#
1548022 and CID#
1548075.
(cherry picked from commit
f7012a93a7f04fa29c7933a4963aa17fcf120e97)
Frantisek Sumsal [Thu, 1 Aug 2024 11:14:52 +0000 (13:14 +0200)]
test: attempt to install sshd-session from multiple places
On Fedora the sshd-session binary is under /usr/libexec/openssh/ so
cover this path as well in the old framework.
Follow-up for
aaa7b36bd15ca3a96a1e11a557482b0bc59c769f.
(cherry picked from commit
ce2344bbee468c7d524e508a8ff7f5f7ae85acb5)
Daan De Meyer [Wed, 31 Jul 2024 15:42:56 +0000 (17:42 +0200)]
mkosi: Beef up testuser a bit
Give it a password and add it to some common groups.
(cherry picked from commit
3fe25d4530356b3b04d505bc5442503897c124dd)
Michal Sekletar [Tue, 30 Jul 2024 14:22:03 +0000 (16:22 +0200)]
socket: fix socket activation of stopped services with pinned FD store
(cherry picked from commit
941a12dcba57f6673230a9c413738c51374d2998)
Daan De Meyer [Wed, 31 Jul 2024 13:02:07 +0000 (15:02 +0200)]
exec-credential: Skip duplicate credentials in load_credential_glob()
We document that when multiple credentials of the same name are found,
we use the first one found so let's actually implement that behavior.
(cherry picked from commit
3de13e6148731ae9c36885afd78b1421e6f16305)
Daan De Meyer [Wed, 31 Jul 2024 13:01:40 +0000 (15:01 +0200)]
exec-credential: Log if we skip duplicate credential
(cherry picked from commit
590348e2bf8415053487324d47d0083b49dfdeb0)
Daan De Meyer [Wed, 31 Jul 2024 13:21:17 +0000 (15:21 +0200)]
Drop EEXIST handling in load_credential_glob()
Credentials are written to a temporary file and renamed to the
destination with renameat() which will replace existing files so
EEXIST should not happen so drop the handling for EEXIST.
(cherry picked from commit
2c2ed3272b992b6eb13864e49847c943e42be00f)
Daan De Meyer [Wed, 31 Jul 2024 11:38:50 +0000 (13:38 +0200)]
cgroup-util: Don't try to open pidfd for kernel threads
The kernel might start returning -EINVAL when trying to open pidfd's
for kernel threads so let's not try to open pidfd's for kernel threads.
(cherry picked from commit
ead48ec35c863650944352a3455f26ce3b393058)
Daan De Meyer [Wed, 31 Jul 2024 09:39:04 +0000 (11:39 +0200)]
log: Fix size calculation for number of iovecs
Each log context field can expand to up to three iovecs (key, value
and newline) so let's fix the size calculation to take this into
account.
(cherry picked from commit
fc83ff3f55ee53fd9101d4e45736f3f996ee7ca6)
Daan De Meyer [Wed, 31 Jul 2024 11:11:51 +0000 (13:11 +0200)]
execute: Drop log level to unit log level in exec_spawn()
All messages logged from exec_spawn() are attributed to the unit
and as such we should set the log level to the unit's max log level
for the duration of the function.
(cherry picked from commit
7881f485c9f57b1c7de4308eeab54458890c5c19)
Dan Nicholson [Tue, 30 Jul 2024 13:37:40 +0000 (07:37 -0600)]
firstboot: fix root params with creds and prompting disabled
Remove an early return that prevents --prompt-root-password or
--prompt-root-shell and systemd.firstboot=off using credentials. In that case,
arg_prompt_root_password and arg_prompt_root_shell will be false, but the
prompt helpers still need to be called to read the credentials. Furthermore, if
only the root shell has been set, don't overwrite the root password.
(cherry picked from commit
35bc4c34240afdd55e117b909f26fa9a5dc54f3b)
Dan Nicholson [Tue, 30 Jul 2024 19:42:26 +0000 (13:42 -0600)]
firstboot: handle missing root password entries
If /etc/passwd and/or /etc/shadow exist but don't have an existing root entry,
one needs to be added. Previously this only worked if the files didn't exist.
(cherry picked from commit
2319154a6bec7b8c42e901dfacaefe95bf4e3750)
Yu Watanabe [Sun, 23 Jun 2024 20:03:58 +0000 (05:03 +0900)]
man: suggest to enable global IPv6Forwarding= setting to make IPv6 packets forwarded
Closes #33414.
(cherry picked from commit
175cdefd33788eced7fd6d76fb0b1676eb6ba799)
Dan Nicholson [Tue, 30 Jul 2024 17:11:11 +0000 (11:11 -0600)]
firstboot: create locked and empty root passwords consistently
Although locked and empty passwords in /etc/passwd are treated the same, in all
other cases the entry is configured to read the password from /etc/shadow.
(cherry picked from commit
5088de9daa156a095e79684c658f9035db971538)
Dan Nicholson [Wed, 31 Jul 2024 00:20:13 +0000 (18:20 -0600)]
test: extend firstboot testing
Several features were not being tested or weren't being evaluated thoroughly.
(cherry picked from commit
38688bbc8ffb16a449a41cab344c27f6b1e74cd3)
Yu Watanabe [Tue, 30 Jul 2024 22:21:09 +0000 (07:21 +0900)]
network: mention that IPv4 ACD is enabled by default for 169.254.0.0/16
Prompted by #33824.
(cherry picked from commit
dd87ad8e193092f053ca702b303b25d2001bf96d)
Yu Watanabe [Tue, 30 Jul 2024 21:49:32 +0000 (06:49 +0900)]
network: request non-NULL SSID when a wlan interface is configured as station
To avoid conflicts with user .network file for the wlan interface with Bond=.
See https://github.com/systemd/systemd/issues/19832#issuecomment-
857661200.
(cherry picked from commit
e2becab08506d8a085f4c18231c7f354db16df9f)
Luca Boccassi [Wed, 31 Jul 2024 00:46:58 +0000 (01:46 +0100)]
efi: fix link to legacy EFI handover protocol
(cherry picked from commit
4d6ab7e8440845301c90211beb22015e7232faa1)
Luca Boccassi [Wed, 31 Jul 2024 00:45:06 +0000 (01:45 +0100)]
stub: allocate and zero enough space in legacy x86 handover protocol
A PE image's memory footprint might be larger than its file size due
to uninitialized memory sections. Normally all PE headers should be
parsed to check the actual required size, but the legacy EFI handover
protocol is only used for x86 Linux bzImages, so we know only the last
section will require extra memory. Use SizeOfImage from the PE header
and if it is larger than the file size, allocate and zero extra memory
before using it.
Fixes https://github.com/systemd/systemd/issues/33816
(cherry picked from commit
19812661f1f65ebe777d1626b5abf6475faababc)
Yu Watanabe [Tue, 30 Jul 2024 18:50:55 +0000 (03:50 +0900)]
network: call link_handle_bound_by_list() before trying to reconfigure interface
Otherwise, when an interface gained its carrier, the interface may not
have matching .network file yet, then link_reconfigure_impl() returns
zero, and link_handle_bound_by_list() is skipped.
Fixes #33837.
(cherry picked from commit
36b8ad085c6902631ad7054bffbda33d6d168823)
Ivan Shapovalov [Sun, 16 Jun 2024 08:46:34 +0000 (10:46 +0200)]
Revert "network/ndisc: ignore most fields of RA header when lifetime is zero"
This reverts commit
ffef01acddcac27caaef02f7f28bff03d7799e5e.
Similar to
2d393b1b6d8 ("network: IPv6 Compliance: Router Advertisement
Processing, Reachable Time [v6LC.2.2.15]"),
Extract from: https://www.ietf.org/rfc/rfc4861.html#section-4.2, p.21,
first paragraph:
The Router Lifetime applies only to
the router's usefulness as a default router; it
does not apply to information contained in other
message fields or options.
So it does not make sense to prevent DHCPv6 when Router Lifetime is 0.
Fixes #33357.
(cherry picked from commit
7767896d127264f660e1e8a714e5cd760840921f)
Daniel P. Berrangé [Tue, 30 Jul 2024 09:51:21 +0000 (10:51 +0100)]
Fix detection of TDX confidential VM on Azure platform
The original CVM detection logic for TDX assumes that the guest can see
the standard TDX CPUID leaf. This was true in Azure when this code was
originally written, however, current Azure now blocks that leaf in the
paravisor. Instead it is required to use the same Azure specific CPUID
leaf that is used for SEV-SNP detection, which reports the VM isolation
type.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
(cherry picked from commit
9d7be044cad1ae54e344daf8f2ec37da46faf0fd)
Yu Watanabe [Tue, 30 Jul 2024 18:04:04 +0000 (03:04 +0900)]
network: do not bring down bound interfaces immediately
Even if a timespan specified to IgnoreCarrierLoss= for an interface,
when the carrier of the interface lost, bound interfaces might be bring
down immediately.
Let's also postpone bringing down bound interfaces with the specified
timespan.
(cherry picked from commit
e8eaed0240d642e70c567b08f3593e4cf45a255a)