1
0
mirror of https://github.com/systemd/systemd.git synced 2024-12-23 21:35:11 +03:00
Commit Graph

64265 Commits

Author SHA1 Message Date
Yu Watanabe
3092eaa72e sd-journal: add _OBJECT_TYPE_INVALID as usual 2023-05-01 14:25:38 +09:00
Yu Watanabe
deb87cc80e sd-journal: align table 2023-05-01 14:25:38 +09:00
Yu Watanabe
5b895f56c6 sd-journal: check that the journal file is not stored in .newest_by_boot_id on free 2023-05-01 14:24:55 +09:00
Yu Watanabe
f11de491ca sd-journal: unset prioq index on failure
Otherwise, potentially, the assertion in
journal_file_unlink_newest_by_bood_id() will be triggered.
2023-05-01 14:24:55 +09:00
Yu Watanabe
e28d825127 sd-journal: fix use-after-free
As commented in the code, we need to replace the pointer to the key,
hence, hashmap_replace() must be used, instead of hashmap_update().

Fixes #27459.
2023-05-01 14:24:55 +09:00
Dan Streetman
9589824158 tpm2: move openssl-required ifdef code out of policy-building function 2023-05-01 00:52:28 +01:00
don bright
46b8c3f5b2
hwdb: add hardware rfkill key for Dell Latitude E6* models (#27462)
Hello

This pull req is adapting pull req #5772 (which fixed issue #5047), for the very similar computer Dell Latitude E6420 which has the same problem with the hardware switch to toggle wifi (aka rfkill). The symptom is the following repeated msgs in dmesg

[  309.010284] atkbd serio0: Use 'setkeycodes e008 <keycode>' to make it known.
[  309.016020] atkbd serio0: Unknown key pressed (translated set 2, code 0x88 on isa0060/serio0).

Adding this line to include E6 models causes these messages to stop showing in dmesg

Thank you
2023-04-30 11:33:13 +08:00
Brad Fitzpatrick
cd18d98349 man: clarify RoutingPolicyRule.TypeOfService docs 2023-04-30 04:30:35 +09:00
Yu Watanabe
ebd6384d57
Merge pull request #27461 from bluca/coredumpctl_completion
coredumpctl: fix bash completion
2023-04-30 04:28:45 +09:00
Luca Boccassi
448ed94dd9 coredumpctl: add --file/--root/--image to bash completion 2023-04-29 18:47:34 +01:00
Luca Boccassi
20ed583e35 coredumpctl: fix bash completion matching
When multi-word matching string is quoted, __contains_word compares
it as a whole to the passed option, so it doesn't work.
2023-04-29 18:47:34 +01:00
Frantisek Sumsal
992e06df08 test: make sure the test units have a test- prefix
So when collecting coverage the test-.service dropin works as
expected.

Follow-up to 969f5f3cea.
2023-04-29 16:59:08 +02:00
Frantisek Sumsal
edea0d6ac5 test: match all messages with the FILE field
As the systemd-pstore process is quite short lived, it might sometimes
lack the necessary metadata to make matching against a unit or a syslog
tag work. Since we already use a cursor file to make the matching window
small as possible, let's just drop the unit match completely and hope
for the best.

Resolves: #27453
2023-04-29 16:58:40 +02:00
Luca Boccassi
708d63c0e7
Merge pull request #27451 from yuwata/core-path-trigger-notify
core/path: do not install new job in .trigger_notify()
2023-04-29 12:35:24 +01:00
Yu Watanabe
8c59feed5e missing_fs: also define struct file_clone_range
Follow-up for b640e274a7.

Addresses b640e274a7 (r110996661).
2023-04-29 07:03:09 +02:00
Yu Watanabe
843cb225ad test: add tests for "systemctl stop" vs triggering by path unit 2023-04-29 09:10:28 +09:00
Yu Watanabe
2e701a7946 test: create temporary units under /run 2023-04-29 09:10:28 +09:00
Yu Watanabe
bc6377762c core/path: do not enqueue new job in .trigger_notify callback
Otherwise,
1. X.path triggered X.service, and the service has waiting start job,
2. systemctl stop X.service
3. the waiting start job is cancelled to install new stop job,
4. path_trigger_notify() is called, and may reinstall new start job,
5. the stop job cannot be installed, and triggeres assertion.

So, instead, let's add a defer event source, then enqueue the new start
job after the stop (or any other type) job finished.

Fixes https://github.com/systemd/systemd/issues/24577#issuecomment-1522628906.
2023-04-29 08:54:29 +09:00
Yu Watanabe
34a4e22e97 core/path: align table 2023-04-29 08:45:06 +09:00
Lennart Poettering
c23e9b6a64 pid1: unify implemenation of /run/ disk space safety check a bit
reload/reexec currently used a separate implementation of the /run/ disk
space check, different from the one used for switch-root, even though
the code is mostly the same. The one difference is that the former
checks are authoritative, the latter are just informational (that's
because refusing a reload/reexec is relatively benign, but refusing a
switch-root quite troublesome, since this code is entered when it's
already "too late" to turn turn back, i.e. when the preparatory
transaction to initiate the switch root are already fully executed.

Let's share some code, and unify codepaths.

(This is preparation for later addition of a "userspace reboot" concept)

No change in behaviour, just refactoring.
2023-04-28 23:27:20 +01:00
Lennart Poettering
5ae89ef347 core/systemctl: when switching root default to /sysroot/
We hardcode the path the initrd uses to prepare the final mount point at
so many places, let's also imply it in "systemctl switch-root" if not
specified.

This adds the fallback both to systemctl and to PID 1 (this is because
both to — different – checks on the path).
2023-04-28 23:26:20 +01:00
Daan De Meyer
77b7026668 libsystemd: Add missing memory pressure functions to public symbols 2023-04-28 21:19:38 +02:00
Mike Yuan
0102857b52
Merge pull request #27445 from poettering/reflink-fix
remove duplication reflink ioctl invocation
2023-04-29 01:33:21 +08:00
Lennart Poettering
4b7fda8731 systemctl: rework 'if' to 'switch' statement 2023-04-29 01:23:24 +08:00
Luca Boccassi
398d858d1c
Merge pull request #27437 from mrc0mmand/pstore
test: add a couple of tests for systemd-pstore
2023-04-28 16:14:09 +01:00
Lennart Poettering
535358ad2e copy: shortcut reflink_range() to reflink() in some cases 2023-04-28 15:16:30 +02:00
Lennart Poettering
427ad121c2 copy: don't call clone ioctls twice
The btrfs name and the generic name have the same values, hence there's
no point in bothering with the former.
2023-04-28 15:10:24 +02:00
Frantisek Sumsal
6858e32d73 test: add a couple of tests for systemd-pstore 2023-04-28 13:16:45 +02:00
Luca Boccassi
91010bd6cc
Merge pull request #27440 from yuwata/reflink-follow-ups
copy: follow ups for reflink()
2023-04-28 11:34:41 +01:00
Frantisek Sumsal
ad5980803a pstore: avoid opening the dmesg.txt file if not requested
Even with Storage=journal we would still attempt to open the final
dmesg.txt file which causes a lot of noise in the journal:

```
[    5.764111] H testsuite-82.sh[658]: + systemctl start systemd-pstore
[    5.806385] H systemd[1]: Starting modprobe@efi_pstore.service...
[    5.808656] H systemd[1]: modprobe@efi_pstore.service: Deactivated successfully.
[    5.808971] H systemd[1]: Finished modprobe@efi_pstore.service.
[    5.818845] H kernel: audit: type=1130 audit(1682630623.637:114): pid=1 uid=0 auid=4294967295 ses=4294967295 subj=kernel msg='unit=modprobe@efi_pstore comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? termin>
[    5.818865] H kernel: audit: type=1131 audit(1682630623.637:115): pid=1 uid=0 auid=4294967295 ses=4294967295 subj=kernel msg='unit=modprobe@efi_pstore comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? termin>
[    5.816052] H systemd[1]: Starting systemd-pstore.service...
[    5.840703] H systemd-pstore[806]: PStore dmesg-efi-168263062313014.
[    5.841239] H systemd-pstore[806]: Failed to open file /var/lib/systemd/pstore/1682630623/014/dmesg.txt: Operation not permitted
[    5.841428] H systemd-pstore[806]: PStore dmesg-efi-168263062312014.
[    5.841575] H systemd-pstore[806]: Failed to open file /var/lib/systemd/pstore/1682630623/014/dmesg.txt: Operation not permitted
[    5.841712] H systemd-pstore[806]: PStore dmesg-efi-168263062311014.
[    5.841839] H systemd-pstore[806]: Failed to open file /var/lib/systemd/pstore/1682630623/014/dmesg.txt: Operation not permitted
[    5.841989] H systemd-pstore[806]: PStore dmesg-efi-168263062310014.
[    5.842141] H systemd-pstore[806]: Failed to open file /var/lib/systemd/pstore/1682630623/014/dmesg.txt: Operation not permitted
[    5.842274] H systemd-pstore[806]: PStore dmesg-efi-168263062309014.
[    5.842423] H systemd-pstore[806]: Failed to open file /var/lib/systemd/pstore/1682630623/014/dmesg.txt: Operation not permitted
[    5.842589] H systemd-pstore[806]: PStore dmesg-efi-168263062308014.
[    5.842722] H systemd-pstore[806]: Failed to open file /var/lib/systemd/pstore/1682630623/014/dmesg.txt: Operation not permitted
[    5.842865] H systemd-pstore[806]: PStore dmesg-efi-168263062307014.
[    5.843003] H systemd-pstore[806]: Failed to open file /var/lib/systemd/pstore/1682630623/014/dmesg.txt: Operation not permitted
[    5.843153] H systemd-pstore[806]: PStore dmesg-efi-168263062306014.
[    5.843280] H systemd-pstore[806]: Failed to open file /var/lib/systemd/pstore/1682630623/014/dmesg.txt: Operation not permitted
[    5.843434] H systemd-pstore[806]: PStore dmesg-efi-168263062305014.
[    5.843570] H systemd-pstore[806]: Failed to open file /var/lib/systemd/pstore/1682630623/014/dmesg.txt: Operation not permitted
[    5.843702] H systemd-pstore[806]: PStore dmesg-efi-168263062304014.
[    5.843831] H systemd-pstore[806]: Failed to open file /var/lib/systemd/pstore/1682630623/014/dmesg.txt: Operation not permitted
[    5.843958] H systemd-pstore[806]: PStore dmesg-efi-168263062303014.
[    5.844093] H systemd-pstore[806]: Failed to open file /var/lib/systemd/pstore/1682630623/014/dmesg.txt: Operation not permitted
[    5.844250] H systemd-pstore[806]: PStore dmesg-efi-168263062302014.
[    5.844412] H systemd-pstore[806]: Failed to open file /var/lib/systemd/pstore/1682630623/014/dmesg.txt: Operation not permitted
[    5.844619] H systemd-pstore[806]: PStore dmesg-efi-168263062301014.
[    5.844781] H systemd-pstore[806]: Failed to open file /var/lib/systemd/pstore/1682630623/014/dmesg.txt: Operation not permitted
[    5.844956] H systemd-pstore[806]: PStore dmesg-efi-168263062300014.
[    5.845168] H systemd-pstore[806]: Failed to open file /var/lib/systemd/pstore/1682630623/014/dmesg.txt: Operation not permitted
[    5.851101] H systemd[1]: Finished systemd-pstore.service.
```
2023-04-28 12:00:42 +02:00
Yu Watanabe
657152a459 network-generator: shorten code a bit 2023-04-28 18:50:47 +09:00
Yu Watanabe
75fd8ad008
Merge pull request #27424 from dtardon/auto-cleanup
More automatic cleanup
2023-04-28 18:46:36 +09:00
Frantisek Sumsal
a95d96a243 pstore: explicitly set the base when converting record ID 2023-04-28 11:41:52 +02:00
David Tardon
360179ea46 logind-session-dbus: use _cleanup_ 2023-04-28 09:29:04 +02:00
David Tardon
80e52052db logind-session-device: modernize session_device_free() 2023-04-28 09:29:03 +02:00
David Tardon
fc447921a8 homed-manager-bus: use _cleanup_ 2023-04-28 09:29:03 +02:00
David Tardon
7fb1cc85ab transaction: make transaction_free() static
... as it's not needed outside transaction.c anymore.
2023-04-28 09:28:25 +02:00
David Tardon
32620826cf manager: use _cleanup_ in manager_propagate_reload() 2023-04-28 09:27:37 +02:00
David Tardon
c7e3eb5dd5 manager: use _cleanup_ in manager_add_job() 2023-04-28 09:27:37 +02:00
David Tardon
542fe40845 transaction: introduce transaction_abort_and_free()
Will be used in following commits.
2023-04-28 09:26:20 +02:00
Yu Watanabe
d6f96778b1 missing_fs: mention commit hash and version of ioctl commands introduced 2023-04-28 14:18:48 +09:00
Yu Watanabe
71e84b4be6 copy: rename reflink_full() -> reflink_range()
The commit b640e274a7 introduced reflink()
and reflink_full(). We usually name function xyz_full() for fully
parameterized version of xyz(), and xyz() is typically a inline alias of
xyz_full(). But in this case, reflink() and reflink_full() call
different ioctl().
Moreover, reflink_full() does partial reflink, while reflink() does full
file reflink. That's super confusing.
Let's rename reflink_full() to reflink_range(), the new name is
consistent with ioctl name, and should be fine.
2023-04-28 14:06:00 +09:00
David Edmundson
c63dde8099 xdg-autostart-service: handle gnome autostart phase better on other desktops
Autostart files which contain the line gnome-autostart-phase are currently
completely skipped by systemd. This is because these are handled internally by
gnome startup through other means.

The problem is a number of desktop files that need to run on KDE too have this
flag set. Ideally they should just create systemd user units, but we're not at
this point universally yet.

This patch changes the logic so if the flag is set, we set NotShowIn-gnome,
which in turn would just not load decided at runtime.

As an optimisation if we would get conflicting OnlyShowIn lines we still
skip the file completely.

Example:
  $ rg 'Exec|Autostart-Phase' /etc/xdg/autostart/gnome-keyring-pkcs11.desktop
  Exec=/usr/bin/gnome-keyring-daemon --start --components=pkcs11
  X-GNOME-Autostart-Phase=PreDisplayServer

  $ cat '/tmp/xxx/app-gnome\x2dkeyring\x2dpkcs11@autostart.service'
  # Automatically generated by systemd-xdg-autostart-generator
  [Unit]
  SourcePath=/etc/xdg/autostart/gnome-keyring-pkcs11.desktop
  ...
  [Service]
  ...
  ExecCondition=/usr/lib/systemd/systemd-xdg-autostart-condition "Unity:MATE" "GNOME"

Co-authored-by: Zbigniew Jędrzejewski-Szmek <zbyszek@in.waw.pl>
2023-04-28 12:32:53 +08:00
Daan De Meyer
b640e274a7 copy: Introduce reflink() and reflink_full()
The kernel has had filesystem independent reflink ioctls for a
while now, let's try to use them and fall back to the btrfs specific
ones if they're not supported.
2023-04-28 00:57:03 +01:00
David Tardon
12da8805e7 transaction: modernize transaction_free() 2023-04-27 21:42:38 +02:00
David Tardon
0a956e8e02 initctl: use _cleanup_ 2023-04-27 21:42:38 +02:00
David Tardon
70bbf65942 initctl: add assert 2023-04-27 21:42:38 +02:00
David Tardon
1cc3a1302f specifier: use _cleanup_ 2023-04-27 21:42:38 +02:00
David Tardon
3cbbefc2a7 path-util: use _cleanup_ 2023-04-27 21:42:38 +02:00
David Tardon
5ce8ce908d logind: rename function to avoid confusion 2023-04-27 21:42:38 +02:00