]>
git.ipfire.org Git - thirdparty/systemd.git/log
Daan De Meyer [Thu, 18 Apr 2024 11:27:05 +0000 (13:27 +0200)]
mkosi: Drop systemd-repart from package lists
This is just a Provides for systemd-udev.
Daan De Meyer [Thu, 18 Apr 2024 11:26:44 +0000 (13:26 +0200)]
mkosi: Update to latest
Luca Boccassi [Mon, 8 Apr 2024 00:57:26 +0000 (01:57 +0100)]
bash completion: add missing parameters for portablectl
Luca Boccassi [Mon, 8 Apr 2024 00:34:12 +0000 (01:34 +0100)]
portablectl: add --clean parameter for detaching
Calls CleanUnit on each portable service being removed, after it has
stopped
Luca Boccassi [Thu, 18 Apr 2024 09:46:19 +0000 (10:46 +0100)]
NEWS: mention GNOME Foundation in contributors list
Sponsored work on homed
Lennart Poettering [Thu, 18 Apr 2024 09:18:35 +0000 (11:18 +0200)]
systemctl: show invocation ID in unit status output
I think we should put more emphasis on the invocation ID as a handle for
a specific runtime cycle of a unit. Let's start with actually showing it
to users.
See: #16035
Jonathan Conder [Wed, 17 Apr 2024 20:01:27 +0000 (08:01 +1200)]
man: document other keyname options for pam_systemd_loadkey
Luca Boccassi [Thu, 18 Apr 2024 08:55:01 +0000 (10:55 +0200)]
Merge pull request #32324 from mrc0mmand/more-website-fixes
docs: use absolute links for our pages
Lennart Poettering [Wed, 17 Apr 2024 19:56:41 +0000 (21:56 +0200)]
doc: fix .ssh credential examples
Let's create the .ssh dir with the right perms first.
Suggested by @gcb.
Fixes: #28172
Daan De Meyer [Wed, 17 Apr 2024 21:38:14 +0000 (23:38 +0200)]
docs: Add note on packages produced by mkosi builds
Norbert Lange [Tue, 26 Mar 2024 23:41:41 +0000 (00:41 +0100)]
openssl-util: compatible with restricted openssl3
openssl can be built without support for engines or with deprecated
definitions disabled.
This also will not pull in most headers automatically,
so add the rsa.h and ec.h header explicitly.
Remove Engine stuff from the header - it is only needed in one
source file.
Make Engine support dependent on the macros.
Norbert Lange [Tue, 26 Mar 2024 23:29:14 +0000 (00:29 +0100)]
resolved-dnstls: remove deprecated openssl functions
There are replacements available in OpenSLL 1.1.0,
but those should not be needed. To quote the docs:
> As of version 1.1.0 OpenSSL will automatically allocate all resources
> that it needs so no explicit initialisation is required.
> Similarly it will also automatically deinitialise as required.
Yu Watanabe [Thu, 18 Apr 2024 05:25:52 +0000 (14:25 +0900)]
udev: fix assignment of ret_truncated
Follow-ups for
089bef66316e5bdc91b9984148e5a6455449c1da .
Yu Watanabe [Thu, 18 Apr 2024 04:08:09 +0000 (13:08 +0900)]
Yu Watanabe [Thu, 18 Apr 2024 02:38:48 +0000 (11:38 +0900)]
Merge pull request #32192 from yuwata/part-scan
blockdev-util: fix detection of partscan
Yu Watanabe [Thu, 18 Apr 2024 01:26:07 +0000 (10:26 +0900)]
Merge pull request #32249 from CodethinkLabs/vmspawn/predicatable_tap_names
vmspawn: generate predicatable TAP device names and MAC addresses
Yu Watanabe [Mon, 8 Apr 2024 02:57:42 +0000 (11:57 +0900)]
blockdev-util: also read 'ext_range' sysattr to check if the partscan is enabled
The 'capability' sysattr was deprecated by
https://github.com/torvalds/linux/commit/
e81cd5a983bb35dabd38ee472cf3fea1c63e0f23 (v6.3).
Yu Watanabe [Mon, 8 Apr 2024 02:56:58 +0000 (11:56 +0900)]
sd-device: introduce device_get_sysattr_unsigned_full()
Yu Watanabe [Thu, 18 Apr 2024 00:45:51 +0000 (09:45 +0900)]
man: fix wrong version info (#31949)
Fixes #31920.
MaxHearnden [Thu, 18 Apr 2024 00:44:22 +0000 (01:44 +0100)]
systemctl: allow user to suppress output when no action scheduled (#32278)
Winterhuman [Thu, 18 Apr 2024 00:43:28 +0000 (00:43 +0000)]
journalctl: update help to say "priority range" (#32323)
Clarify that `-p, --priority=` always treats its option as a priority range, even when given
a single log level per the full man page description.
Co-authored-by: Mike Yuan <me@yhndnzj.com>
Yu Watanabe [Tue, 16 Apr 2024 08:47:18 +0000 (17:47 +0900)]
test-network: add test case of RS sent by the same interface
Yu Watanabe [Tue, 16 Apr 2024 08:45:15 +0000 (17:45 +0900)]
test-network: introduce check_networkd_log() helper function
Yu Watanabe [Tue, 16 Apr 2024 08:44:17 +0000 (17:44 +0900)]
test-network: drop trailing spaces
Follow-up for
fb573007430ab0dbe45517b58837d2fa5cfa1a48 .
Yu Watanabe [Tue, 16 Apr 2024 07:28:44 +0000 (16:28 +0900)]
network/ndisc: disable Neighbor discovery client if RADV is enabled
Running both sd-ndisc and sd-radv should be mostly a misconfiguration,
but may not. So, let's only disable sd-ndisc by default when sd-radv is
enabled, but allow when both are explicitly requested.
Yu Watanabe [Tue, 16 Apr 2024 07:23:36 +0000 (16:23 +0900)]
sd-radv: refuse packet from the same interface
Prompted by https://github.com/systemd/systemd/pull/32267#discussion_r1566721306.
Yu Watanabe [Thu, 18 Apr 2024 00:28:16 +0000 (09:28 +0900)]
Merge pull request #32308 from yuwata/sd-ndisc-disable-timer-on-recv
sd-ndisc: disable one more timer event source
Yu Watanabe [Wed, 17 Apr 2024 21:18:09 +0000 (06:18 +0900)]
Merge pull request #32318 from YHNdnzj/skip-leading-char
journal-gatewayd: use skip_leading_chars
Jonathan Conder [Wed, 17 Apr 2024 19:56:52 +0000 (07:56 +1200)]
man: add pam_gnome_keyring to auth section after pam_systemd_loadkey
This is required because pam_sm_open_session [1] only looks at
gkr_system_authtok, which is copied from the kernel keyring in
pam_sm_authenticate.
[1] https://gitlab.gnome.org/GNOME/gnome-keyring/-/blob/46.1/pam/gkr-pam-module.c?ref_type=tags
Lennart Poettering [Wed, 17 Apr 2024 20:12:49 +0000 (22:12 +0200)]
Merge pull request #32320 from bluca/softreboot_serialize
Soft reboot timestamp follow-ups
Zbigniew Jędrzejewski-Szmek [Wed, 17 Apr 2024 16:11:31 +0000 (18:11 +0200)]
man: mention that sd_journal_test_cursor() needs a positioning call
Fixes #30331.
Daan De Meyer [Wed, 17 Apr 2024 19:29:28 +0000 (21:29 +0200)]
Merge pull request #32317 from DaanDeMeyer/mkosi
mkosi: Set up -ffile-prefix-map= correctly when building debuginfo packages
Frantisek Sumsal [Wed, 17 Apr 2024 18:54:45 +0000 (20:54 +0200)]
docs: use absolute links for our pages
Since
56b2970 has proven to be a no-go for us, as it breaks existing
links, let's embrace the trailing slash and use absolute links
everywhere for our pages. This way we'll get around browser cleverly
appending the relative link to the current location (since it ends with
a slash), and given our docs/ layout is flat it's not much of a hassle
either.
Converted using this beauty:
$ sed -ri 's/(\[.+\]\()([A-Z_]+\))/\1\/\2/g' *.md
Resolves: #32088 (again) and #32310
Frantisek Sumsal [Wed, 17 Apr 2024 18:42:58 +0000 (20:42 +0200)]
Revert "docs: don't suffix page permalink with a slash"
This breaks existing links with a trailing slash, so it's a no-go.
This reverts commit
d1a7e030c5daa4a09ee93b8af6b28ecaac5d34d0 .
Luca Boccassi [Tue, 16 Apr 2024 14:33:09 +0000 (15:33 +0100)]
core: rename SoftRebootStartTimestamp -> ShutdownStartTimestamp and generalize
Follow-up for
54f86b86ba8168faccbfc2ad16ceeccefde98a51
Luca Boccassi [Thu, 19 Oct 2023 23:34:29 +0000 (00:34 +0100)]
manager: serialize current objective
So that we can tell what happened before the exec. It is overwritten
shortly after deserialization. Use it to bump the soft reboots counter.
Lennart Poettering [Wed, 17 Apr 2024 17:18:44 +0000 (19:18 +0200)]
Merge pull request #32315 from poettering/hwdb-256
hwdb/autosuspend/NEWS update for 256-rc1
Zbigniew Jędrzejewski-Szmek [Wed, 17 Apr 2024 15:34:43 +0000 (17:34 +0200)]
Merge pull request #31648 from neighbourhoodie/review-content
I did the merge manually to resolve a trivial conflict.
Daan De Meyer [Wed, 17 Apr 2024 14:59:08 +0000 (16:59 +0200)]
mkosi: Drop workarounds
With the latest Github Actions image release none of these should
be required anymore (
20240414 .1.1).
Daan De Meyer [Wed, 17 Apr 2024 14:10:26 +0000 (16:10 +0200)]
mkosi: Switch to O0 and disable FORTIFY_SOURCE
-Og still causes a lot of "<optimized out>" in GDB so let's use -O0
instead and disable FORTIFY_SOURCE as it doesn't work without
optimizations enabled.
Lennart Poettering [Wed, 17 Apr 2024 15:03:48 +0000 (17:03 +0200)]
Merge pull request #32143 from yuwata/magic
basic/linux: copy magic.h from kernel to our code
Lennart Poettering [Wed, 17 Apr 2024 09:24:59 +0000 (11:24 +0200)]
NEWS: prepare -rc1
Lennart Poettering [Wed, 17 Apr 2024 10:40:24 +0000 (12:40 +0200)]
autosuspend: update for v256
Lennart Poettering [Wed, 17 Apr 2024 09:22:14 +0000 (11:22 +0200)]
hwdb: update for v256
Lennart Poettering [Wed, 17 Apr 2024 14:20:51 +0000 (16:20 +0200)]
Merge pull request #32311 from YHNdnzj/write-ellipsis
Some minor cleanup for string-util (ellipsis)
Daan De Meyer [Wed, 17 Apr 2024 13:54:53 +0000 (15:54 +0200)]
mkosi: Disable DWARF debug info optimization
We care more about fast builds in this case so disable the DWARF
debug optimization that's done by default.
Mike Yuan [Wed, 17 Apr 2024 14:05:37 +0000 (22:05 +0800)]
NEWS: udev and networkd pick up more things from creds
Mike Yuan [Wed, 17 Apr 2024 13:50:22 +0000 (21:50 +0800)]
journal-gatewayd: use skip_leading_chars where appropriate
Prompted by
c5d675472588815dedf67af8c36da0fac398fb3e
Mike Yuan [Wed, 17 Apr 2024 13:50:05 +0000 (21:50 +0800)]
journal-gatewayd: add missing assertions
Daan De Meyer [Wed, 17 Apr 2024 13:38:24 +0000 (15:38 +0200)]
mkosi: Override build_cflags instead of _distro_extra_cflags
The latter is not supported on CentOS Stream 9.
Antonio Alvarez Feijoo [Wed, 17 Apr 2024 09:19:01 +0000 (11:19 +0200)]
journal-gatewayd: remove duplicate call to `startswith()`
Daan De Meyer [Wed, 17 Apr 2024 12:55:08 +0000 (14:55 +0200)]
mkosi: Set up -ffile-prefix-map= correctly when building debuginfo packages
This makes sure that the debuginfo files contain source files pointing
to the source files shipped by the debugsource package.
Normally this should be done automatically by rpm invoking debugedit
but for some unknown reason debugedit refuses to rewrite the source
files in our binaries.
Given that debugedit is completely undebugable (does not generate any
logs at all, and its source code is ridiculously obtuse), let's set
-ffile-prefix-map= when building instead which achieves the same
effect.
dependabot[bot] [Wed, 17 Apr 2024 09:56:56 +0000 (09:56 +0000)]
build(deps): bump pkg/debian from `
e477254 ` to `
30c77a7 `
Bumps pkg/debian from `
e477254 ` to `
30c77a7 `.
---
updated-dependencies:
- dependency-name: pkg/debian
dependency-type: direct:production
...
Signed-off-by: dependabot[bot] <support@github.com>
Mike Yuan [Wed, 17 Apr 2024 08:48:29 +0000 (16:48 +0800)]
core/show-status: use strlen where appropriate
We don't use sizeof() outside of array definitions
these days. The compiler should be capable of
optimizing this out.
Mike Yuan [Wed, 17 Apr 2024 08:48:00 +0000 (16:48 +0800)]
string-util: use mempcpy where appropriate
Yu Watanabe [Wed, 17 Apr 2024 01:01:39 +0000 (10:01 +0900)]
string-util: use special_glyph() at one more place
No functional change, just refactoring.
Yu Watanabe [Tue, 16 Apr 2024 06:17:50 +0000 (15:17 +0900)]
fuzz-ndisc-rs: also test packets with sd-radv
Addresses https://github.com/systemd/systemd/pull/32267#discussion_r1566715787.
Suggested-by: Evgeny Vereshchagin <evvers@ya.ru>
Yu Watanabe [Wed, 17 Apr 2024 04:29:35 +0000 (13:29 +0900)]
Yu Watanabe [Wed, 17 Apr 2024 04:02:26 +0000 (13:02 +0900)]
sd-ndisc: improve debugging logs
Yu Watanabe [Wed, 17 Apr 2024 03:21:44 +0000 (12:21 +0900)]
sd-ndisc: disable one more timer event source
Otherwise, even if we have already received RA, timeout callback will be
called. Currently, networkd mostly does nothing on timeout, hence should
not change any effective behavior.
Follow-up for
e7cb80474d4e580f7cc7c8645e1412e4a59f7588 .
Yu Watanabe [Wed, 17 Apr 2024 02:52:30 +0000 (11:52 +0900)]
Merge pull request #32300 from mrc0mmand/assorted-tweaks
test: split TEST-50-DISSECT into smaller parts
Yu Watanabe [Wed, 17 Apr 2024 02:49:23 +0000 (11:49 +0900)]
Merge pull request #32303 from CodethinkLabs/push_down_disk_partitioning
test-64-udev-storage: Push disk partitioning into the testsuite itself
Bryan Jacobs [Tue, 16 Apr 2024 03:12:22 +0000 (13:12 +1000)]
cryptenroll: Use CTAP2.1 credProtect extension
When enrolling a new FIDO2 token with a client PIN, this tells the authenticator to require the PIN on all uses.
It also collects a PIN before attempting to create a credential.
Works around #31443 in most (not all) scenarios.
Sam Leonard [Tue, 16 Apr 2024 16:02:31 +0000 (17:02 +0100)]
test-64-udev-storage: partition disk inside testcase - testcase_multipath_basic_failover
Sam Leonard [Tue, 16 Apr 2024 15:17:59 +0000 (16:17 +0100)]
test-64-udev-storage: partition disk inside testcase - testcase_virtio_scsi_identically_named_partitions
Sam Leonard [Tue, 16 Apr 2024 12:18:34 +0000 (13:18 +0100)]
test-64-udev-storage: partition disk inside testcase - testcase_long_sysfs_path
Daan De Meyer [Tue, 16 Apr 2024 14:59:41 +0000 (16:59 +0200)]
Merge pull request #32302 from DaanDeMeyer/mkosi
Various mkosi improvements
Lennart Poettering [Tue, 16 Apr 2024 13:45:06 +0000 (15:45 +0200)]
Merge pull request #32236 from poettering/journalctl-list-namespace-followup
journalctl: --list-namespaces followup
Daan De Meyer [Tue, 16 Apr 2024 09:24:56 +0000 (11:24 +0200)]
mkosi: Add support for building debug packages
This allows building debug packages by setting WITH_DEBUG=1. This
slows down the build a lot so it's opt in. We don't yet install
these but can do so in a future commit.
Daan De Meyer [Tue, 16 Apr 2024 11:59:20 +0000 (13:59 +0200)]
mkosi: Add --noclean when running rpmbuild
The entire build environment is ephemeral anyway so everything is
cleaned regardless. By specifying --noclean, we make debugging
easier as the rpm build root can be inspected when using --debug-shell.
Daan De Meyer [Tue, 16 Apr 2024 11:58:07 +0000 (13:58 +0200)]
mkosi: Install which in image
gdb-headless in Fedora is missing a dependency on which:
https://bugzilla.redhat.com/show_bug.cgi?id=
2275274
This happens all the time so let's make sure which is always installed
in the image.
Daan De Meyer [Mon, 15 Apr 2024 20:28:11 +0000 (22:28 +0200)]
Update debugging with vscode section
- We have ssh-generator now, so need for mkosi's Ssh= option anymore.
- By enabling RuntimeBuildSources= by default, we don't need the gdb
config file in the image anymore, since the build and source
directories will be mounted at the expected locations.
Frantisek Sumsal [Mon, 15 Apr 2024 21:28:43 +0000 (23:28 +0200)]
test: rework systemd-sysext test
Let's rework the test a bit, namely:
- condense the code a bit
- drop unnecessary braces around variables
- drop unnecessary explanations around `touch` calls
- drop/rename functions to make the code more self-explanatory
- simplify cleanup a bit
- create R/O bind mounts directly (supported since util-linux 2.27)
Frantisek Sumsal [Mon, 15 Apr 2024 18:52:59 +0000 (20:52 +0200)]
test: simplify unit check
Frantisek Sumsal [Mon, 15 Apr 2024 18:27:32 +0000 (20:27 +0200)]
test: split some overly long lines
Let's get the line count back up again by splitting some overly long
lines in hopes to make the code slightly more readable.
Frantisek Sumsal [Mon, 15 Apr 2024 17:44:23 +0000 (19:44 +0200)]
test: split TEST-50-DISSECT into smaller parts
As the current version has almost 2k lines.
Lennart Poettering [Thu, 11 Apr 2024 17:07:21 +0000 (19:07 +0200)]
journalctl: tighten rules on parsing namespace journal dir suffixes
The dot must follow the machine ID immediately, let's check for that.
Also, I think it's generally better to parse the machine ID and then
comparing it, instead of comparing the string representation. That's
more in line how we usually do it, as we parse 128bit IDs generally
case-insensitively.
Lennart Poettering [Thu, 11 Apr 2024 17:04:14 +0000 (19:04 +0200)]
journalctl: show human readable text if no namespace is found
The common case will be that no namespaces are defined, hence by default
we'd just output "NAMESPACE" and exit, which is very confusing.
Generate a nice message instead, how many of our tools do it. Suppress
it on --quiet however.
Lennart Poettering [Thu, 11 Apr 2024 17:03:54 +0000 (19:03 +0200)]
journalctl: add missing --help line for --list-namespaces
Follow-up for: #32236
68f66a171398e27280a95e58ae7464219cccaaec
Mike Yuan [Tue, 16 Apr 2024 04:27:23 +0000 (12:27 +0800)]
shared/vpick: also align function params in header
Follow-up for
6d8690ecd34a6c9450e78e242e07b1a4479d6d27
Luca Boccassi [Tue, 16 Apr 2024 08:44:25 +0000 (10:44 +0200)]
Merge pull request #32289 from bluca/counter
soft-reboot counter follow-ups
Yu Watanabe [Tue, 16 Apr 2024 07:42:59 +0000 (16:42 +0900)]
Merge pull request #32294 from yuwata/network-generator-creds
network-generator: also load drop-ins for networkd.conf from credentials
Yu Watanabe [Tue, 16 Apr 2024 07:42:46 +0000 (16:42 +0900)]
Merge pull request #32297 from yuwata/networkctl-cat-without-args
networkctl: allow to call 'networkctl cat' without arguments
Mike Yuan [Tue, 16 Apr 2024 06:22:54 +0000 (14:22 +0800)]
NEWS: a few fixes, add systemd-cat --namespace=
Yu Watanabe [Tue, 16 Apr 2024 01:46:09 +0000 (10:46 +0900)]
network: fix use of wrong flag
Fixes a bug introduced by
86a66e9b95048b1a3a4e297ba2884afcedd1585e (v255).
Yu Watanabe [Tue, 16 Apr 2024 04:30:19 +0000 (13:30 +0900)]
test: add test case for 'networkctl cat' without arguments
Yu Watanabe [Tue, 16 Apr 2024 04:29:23 +0000 (13:29 +0900)]
networkctl: allow to call 'networkctl cat' without arguments
Then, show networkd.conf and its drop-ins.
Yu Watanabe [Tue, 16 Apr 2024 02:50:38 +0000 (11:50 +0900)]
man: add missing drop-in directory
Yu Watanabe [Tue, 16 Apr 2024 03:03:46 +0000 (12:03 +0900)]
test: add test case for network.conf.* credential
Yu Watanabe [Tue, 16 Apr 2024 02:57:23 +0000 (11:57 +0900)]
network-generator: also copy drop-ins for networkd.conf from credential
Follow-up for
1a30285590c2f40f256d0628950ef9243b2c1938 .
Yu Watanabe [Tue, 16 Apr 2024 03:21:36 +0000 (12:21 +0900)]
NEWS: fix typo
Yu Watanabe [Tue, 16 Apr 2024 02:45:22 +0000 (11:45 +0900)]
Merge pull request #32103 from yuwata/udev-load-credentials
udev: load udev rules from credentials
Yu Watanabe [Mon, 15 Apr 2024 02:19:57 +0000 (11:19 +0900)]
sd-radv: drop unnecessary argument
The argument always has the same value.
Yu Watanabe [Sun, 18 Feb 2024 06:23:09 +0000 (15:23 +0900)]
sd-radv: set only basic information on stop
There are many possible reasons to stop the service;
tentative reboot of the service or the system, the router may be
revoked, and so on.
And, each situations, the availability of the previously announced
options e.g. prefixes, DNSSL and so on is not clear.
So, let's announce only the clear information, that is, the router
lifetime is zero. which indicates that the router will be invalidated
(regardless tentative or not).
Yu Watanabe [Tue, 16 Apr 2024 02:18:07 +0000 (11:18 +0900)]
Merge pull request #32267 from yuwata/sd-ndisc-router-solicit
sd-radv: introduce sd_ndisc_router_solicit and use it
Yu Watanabe [Thu, 4 Apr 2024 19:14:19 +0000 (04:14 +0900)]
test: add test for systemd-udev-load-credentials.service
Yu Watanabe [Fri, 23 Feb 2024 06:41:06 +0000 (15:41 +0900)]
sd-radv: use ICMP6Packet and sd_ndisc_router_solicit
Then, sd-radv can reject invalid Router Solicit messages, e.g. that have
invalid options.
No effective functional change, just refactoring.
Yu Watanabe [Fri, 23 Feb 2024 06:19:58 +0000 (15:19 +0900)]
sd-radv: introduce sd_ndisc_router_solicit
It is not used in this stage, but will be used later in sd-radv.
Yu Watanabe [Thu, 4 Apr 2024 19:04:31 +0000 (04:04 +0900)]
units: introduce systemd-udev-load-credentials.service
Yu Watanabe [Thu, 4 Apr 2024 18:55:25 +0000 (03:55 +0900)]
udevadm-control: add --load-credentials option
When specified, credentials udev.conf.* and udev.rules.* are copied to
the corresponding directories.