IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
* a3524fc837 Use a more precise Recommends for libkxbcommon
* 980ede8c0f Drop machined revert
* d569018a92 Rebuilt for the bin-sbin merge
* 8881fa94ee Version 256.2
* 1cc4f83002 Link systemd-executor statically
* 0319e62d9c Update dracut workaround
* c96f54de22 Fix ELN build
* 3f68c5d802 Only exclude dracut conflicts on non-fedora on upstream builds
* 7db154308b Conditionalize dracut Conflicts more
(cherry picked from commit 72a8f508e5)
* 8c025c3bdf Accepting request 1184267 from Base:System
|\
| * 735f8c4ba4 - Import commit 5a8eadd0c021758337a020c423f25a353bdb9b3c (merge of v255.8) For a complete list of changes, visit: 603cd1d4d8...5a8eadd0c0 - Drop 5003-Revert-run-pass-the-pty-slave-fd-to-transient-servic.patch as v255.8 contains the workaround (commit 639c922ede9485) for the broken commit 28459ba1f4.
* | 37853fecc3 Accepting request 1183029 from Base:System
|/
* 638de11012 - Don't automatically clean unmodified config files up (bsc#1226415)
* 369c023c24 reorder one more time...
* ffa9f0ac80 reorder the runtime deps of the testsuite package so the format_spec_file thingy stop screwing up the spec file...
* 12c1190a79 fix rev 1529: the devel packages are really needed by the testsuite script to install the dlopened libs into the image
* ca8e7f54ce - systemd.spec: move a misplaced %endif in the testsuite sub-package.
* b7944f5b14 - Merge systemd-coredump back into the main package (bsc#1091684)
* 3fa0dea84a - Don't pull the devel packages in when installing the testsuite package.
(cherry picked from commit eb6f177ae4)
- Stop installing the policy in the initramfs as it's not really
supported anyway (https://github.com/fedora-selinux/selinux-policy/issues/2221)
- Stop relabeling on first boot and prefer to do it at image build time
- Disable mkosi relabeling by default but enable it in CI
- Build image as root in CI so the SELinux relabeling works properly
(cherry picked from commit dedd712dd9)
stress-ng is available in OpenSUSE and in CentOS Stream without needing
EPEL so let's switch to it instead of stress.
(cherry picked from commit 3e3a15c06e)
Let's make use of mkosi ssh to give everyone a one liner to run instead
of having to execute two commands (one outside the VM, one inside).
(cherry picked from commit 00acdb3147)
Otherwise trying to use --wipe might fail if the build directory
has not been configured but is not empty either.
(cherry picked from commit 73b3179a53)
In https://github.com/systemd/systemd/pull/33659 the commit was
updated to point to my fork without changing it back after the mkosi
PR was merged so let's change it back to point to the official
repository.
(cherry picked from commit c8ce41954b)
In https://github.com/systemd/mkosi/pull/2847, the '@' specifier is
removed, CLI arguments take priority over configuration files again
and the "main" image is defined at the top level instead of in
mkosi.images/. Additionally, not every setting from the top level
configuration is inherited by the images in mkosi.images/ anymore,
only settings which make sense to be inherited are inherited.
This commit gets rid of all the usages of '@', moves the "main" image
configuration from mkosi.images/system to the top level and gets rid
of various hacks we had in place to deal with quirks of the old
configuration parsing logic.
We also remove usages of Images= and --append as these options are
removed by the mentioned PR.
(cherry picked from commit 20345a86b7)
This ensures that even in case the distro repository has newer
versions, the locally built packages are preferred and installed,
even to the point of downgrading already installed ones.
This is needed especially for future stable branches, when the
distros will have a newer version.
(cherry picked from commit 1ee8e526d7)
The contents of this file are passed as arguments to fixfiles so
let's remove the comment as otherwise fixfiles just crashes.
(cherry picked from commit 5eab7c5097)
Let's not insist on btrfs everywhere. 93440db8b5
switched us back to btrfs as we wanted to rely on the fact it records
timestamps properly. Since we now prefer to do incremental builds on the host
with "mkosi -t none" we don't mind anymore that timestamps are not recorded
properly so we're not forced to use btrfs anymore.
This also increases test coverage as we'll now test with different root
filesystems.
(cherry picked from commit 5db8db5e34)
The new link-executor-shared option is similar to the existing
link-udev-shared: when set to false, we link to the static versions of our
internal libraries.
The resulting exuctor binary is fairly large, about as large as libsystemd-core
(14 MB without lto, 8 with lto).
This is intended as a workaround for the fuckup with the pinned executor
binary:
when an upgrade is performed, the package manager will install new version of
the libraries and new version of the code, and some time later reexecute the
managers. This creates a window when the pinned executor binary will fail to
execute. There are two factors which make the issue easier to hit:
- when the distribution uses a finely-grained shared-lib-tag. E.g. Fedora
uses version-release as the tag, which means that the issue occurs on
every package upgrade. This is the right thing to do, because the
ABI of our internal libraries is not stable at all, so replacing the
library from a different version in place creates a window where our
programs may crash or misbehave.
- when the distribution doesn't immediately reexec all the managers after
upgrade. In early versions of systemd, we used to hammer the machine during
upgrade, doing daemon-reexecs repeatedly. This works, but is ugly and
wasteful. Doing the reexecs while the upgrade is in progres also creates a
window where a mix of old and new configs or both is loaded. Users are
particularly annoyed by those reloads if there is some issue in the
configuration causing us to emit warnings on every reexec. Doing the
reexecs once after the new configuration and libraries have been put
in place is nicer.
The pinning of the executor binary breaks upgrades and in particular
it penalizes the distributions which make use of the features which
were previously added to avoid bugs and inefficiency during upgrades.
When the executor is linked statically, there is a smaller chance that it'll
fail to load libraries. The issue can still occur because other libraries, not
our own, are linked dynamically.
(cherry picked from commit d59cae6ceb)
By itself, this is not useful. I'm making this a separate commit to
make debugging easier. It turns out that meson does static libraries
using references, so the "static library" a tiny stub stub that refers
to the object files on disk and this has negligible cost:
$ ls -lhd build/src/core/libsystemd-core-257.{a,so}
-rw-r--r-- 1 zbyszek zbyszek 36K Jul 3 16:54 build/src/core/libsystemd-core-257.a
-rwxr-xr-x 1 zbyszek zbyszek 6.1M Jul 3 16:54 build/src/core/libsystemd-core-257.so
(cherry picked from commit d0689ee5fb)
Our variables for internal libraries are named 'libfoo' for the shared lib
variant, and 'libfoo_static' for the static lib variant. The only exception was
libbasic, because we didn't have a shared variant for it. But let's rename it
for consitency. This makes the build config easier to understand.
(cherry picked from commit 732ed8a84e)
An smbios object with no variable part is a special case, it's just
suffixed with two NUL btes. handle that properly.
This is inspired by a similar fix from https://github.com/systemd/systemd/pull/29726
(cherry picked from commit 44ec70489f)
If the io.systemd.DynamicUser or io.systemd.Machine files exist,
but nothing is listening on them, the nss-systemd module returns
ECONNREFUSED and systemd-sysusers fails to creat the user/group.
This is problematic when ran by packaging scripts, as the package
assumes that after this has run, the user/group exist and can
be used. adduser does not fail in the same situation.
Change sysusers to print a loud warning but otherwise continue
when NSS returns an error.
(cherry picked from commit fc9938d6f8)
The XDG base dir spec adopted ~/.local/state/ as a thing a while back,
and we updated our docs in b4d6bc63e6, but
forgot to to update the table at the bottom to fully reflect the update.
Fix that.
(cherry picked from commit 72a6296b16)
This file doesn't document features of systemd, but is more a of a
general description that generalizes/modernizes FHS. As such, the items
listed in it weren't "added" in systemd versions, they simply reflect
general concepts independent of any specific systemd version. hence
let's drop this misleading and confusing version info.
Or in other words, the man page currently claims under "/usr/": "Added
in version 215." – Which of course is rubbish, the directory existed
since time began.
This also rebreaks all paragaphs this touches.
No content changes.
(cherry picked from commit 26db8fe247)
The previous commit tries to extract a substring from the
extension-release suffix, but that is not right, it's only the
images that need to be versioned and extracted, use the extension-release
suffix as-is. Otherwise if it happens to contain a prefix that
matches the wrong image, it will be taken into account.
Follow-up for 37543971af
(cherry picked from commit 92d1fe3efa)
For MountImages=, if the source is a block device, it will most likely reside
in /dev. It should be also possible to mount a static device file system in
place of (or part of) /dev. So let's allow paths starting with /dev as an
exception for MountImages=.
(cherry picked from commit e81025970f)
Let's mention the new way to install the latest changes without
rebuilding the image. Let's also remove the duplicate info about
distribution packages that is already mentioned in its own section.
(cherry picked from commit a2403af329)
We need to enable this otherwise systemd-oomd.service fails to start.
Fixes:
ConditionControlGroupController=memory was not met
(cherry picked from commit aa329b8922)
The patch is originally from Brenton Simpson, I (Lennart) just added some
comments and rebased it.
I didn't test this, but the patch looks so obviously right to me, that
I think we should just merge it, instead of delaying this further. In
the worst case noone notices, in the best case this makes sd-boot work
reasonably nicely on devices that only have a hadware power key + volume
rocker.
Fixes: #30598
Replaces: #31135
(cherry picked from commit 2fda6f5fff)
Otherwise, busctl --user call ... SoftReboot results in
user manager broadcasting signal and initiating soft-reboot...
(cherry picked from commit 236cd48546)
These are required by the bpf_tracing.h header in libbpf, see
https://github.com/libbpf/libbpf/blob/master/src/bpf_tracing.h.
bpf_tracing.h does have a few fallbacks in case __TARGET_ARCH_XXX
is not defined but recommends using the __TARGET_ARCH macros instead
so let's do that.
(cherry picked from commit 48d6dad100)
We calculate the amount of uncompressed data we can write by taking the limits
into account and halving it to ensure there's room for switching to compression
on the fly when storing cores on a tmpfs (eg: due read-only rootfs).
But the logic is flawed, as taking into account the size of the tmpfs storage
was applied after the halving, so in practice when an uncompressed core file
was larger than the tmpfs, we fill it and then fail.
Rearrange the logic so that the halving is done after taking into account
the tmpfs size.
(cherry picked from commit e6b2508275)
