]>
git.ipfire.org Git - thirdparty/systemd.git/log
Yu Watanabe [Tue, 16 Jan 2024 16:38:55 +0000 (01:38 +0900)]
test-network: test the default required operational state for CAN devices
Yu Watanabe [Tue, 16 Jan 2024 16:35:34 +0000 (01:35 +0900)]
network: make 'carrier' as the default required operational state for CAN device
As CAN devices do not support IP address, hence the state never goes to
higher than 'carrier'.
Prompted by https://github.com/linux-can/can-utils/issues/68#issuecomment-
1327987724 .
Yu Watanabe [Tue, 16 Jan 2024 16:04:18 +0000 (01:04 +0900)]
wait-online: several cleanups for LinkOperationalState
- fix memleak in parser,
- fix missing return in parser on failure,
- drop unnecessary temporary argument in command line argument parser,
- use recently introduced macros and helper functions.
Yu Watanabe [Tue, 16 Jan 2024 16:01:32 +0000 (01:01 +0900)]
network: several cleanups for LinkOperationalState
- introduce link_required_operstate_for_online() helper function,
- use recently introduced macros and helper functions,
- unconditionally serialize the minimum and maximum of required
operational state.
Yu Watanabe [Tue, 16 Jan 2024 15:00:45 +0000 (00:00 +0900)]
network: drop unnecessary temporary variables
Yu Watanabe [Tue, 16 Jan 2024 14:54:39 +0000 (23:54 +0900)]
sd-network: modernize parse_operational_state_range()
- rename 'out' -> 'ret',
- introduce LINK_OPERSTATE_RANGE_INVALID,
- constify LINK_OPERSTATE_RANGE_DEFAULT,
- drop spurious const specifier for allocated string,
- etc,.
Yu Watanabe [Tue, 16 Jan 2024 15:54:34 +0000 (00:54 +0900)]
sd-network: introduce three helper functions for LinkOperationalState
A S Alam [Tue, 16 Jan 2024 14:35:41 +0000 (15:35 +0100)]
po: Translated using Weblate (Punjabi)
Currently translated at 32.1% (73 of 227 strings)
Co-authored-by: A S Alam <aalam@users.noreply.translate.fedoraproject.org>
Translate-URL: https://translate.fedoraproject.org/projects/systemd/master/pa/
Translation: systemd/main
Lennart Poettering [Tue, 16 Jan 2024 17:57:56 +0000 (18:57 +0100)]
Merge pull request #30944 from CodethinkLabs/vmspawn/fix_fedora_issues
vmspawn: fix more issues
Yu Watanabe [Sat, 13 Jan 2024 04:16:10 +0000 (13:16 +0900)]
network/route: find/distinguish routes in the same way that the kernel uses
The kernel uses different logics to find or distinguish IPv4 and IPv6 routes.
Let's follow the same way that the kernel uses.
See comments in the code for more details.
Mike Yuan [Tue, 16 Jan 2024 14:47:11 +0000 (22:47 +0800)]
labeler: add bsod, hibernate-resume, nspawn and vmspawn
Lennart Poettering [Tue, 16 Jan 2024 16:12:00 +0000 (17:12 +0100)]
Merge pull request #30960 from poettering/get-user-creds-modernization
user-util: modernize get_user_creds() + get_group_creds()
Mike Yuan [Tue, 16 Jan 2024 13:27:23 +0000 (21:27 +0800)]
logind: allow Sleep() and CanSleep() dbus calls
Follow-up for
cd4dd90b99ac5617431ce4f0ca1e9089197c85fd
Luca Boccassi [Tue, 16 Jan 2024 15:25:08 +0000 (15:25 +0000)]
Merge pull request #30963 from poettering/creds-tweaks
two minor creds-util.c simplifications
Luca Boccassi [Tue, 16 Jan 2024 15:24:22 +0000 (15:24 +0000)]
Merge pull request #30961 from poettering/varlink-bus-polkit-fixes
varlink polkit client fixes
Luca Boccassi [Tue, 16 Jan 2024 15:18:36 +0000 (15:18 +0000)]
Merge pull request #30959 from poettering/varlink-debug-better
varlink tweaks for making debugging easier
Lennart Poettering [Tue, 16 Jan 2024 13:56:25 +0000 (14:56 +0100)]
user-util: honour PREFER_NSS flag in get_group_creds() the same way as we already do in get_user_creds()
let's onour the PREFER_NSS flag also in get_group_creds() the same
way we honour it in get_user_creds().
Simply to remove surprises.
Lennart Poettering [Mon, 15 Jan 2024 16:00:40 +0000 (17:00 +0100)]
user-util: modernize get_user_creds() + get_group_creds()
Let's rename return parameters to ret_xyz, and let's guarantee the usual
rule to initialize all return parameters on success, and not touch any
on failure.
Lennart Poettering [Mon, 15 Jan 2024 14:21:35 +0000 (15:21 +0100)]
creds-util: drop unnecessary include
Follow-up for:
e653a194e490fae7d166f40762c334006d592051
Lennart Poettering [Mon, 15 Jan 2024 16:33:59 +0000 (17:33 +0100)]
creds-util: simplify offset calculation a bit
Don't recalculate the overall offset entirely each time we process
another header. Instead, if we already validated an earlier offset, just
reuse the result, it's readily available in 'p'.
No change in behaviour, just a bit of code simplification.
Lennart Poettering [Tue, 16 Jan 2024 12:27:32 +0000 (13:27 +0100)]
bus-polkit: don't return positive in varlink_verify_polkit_async() just because we already sent an error reply
A positive return value means "access granted", hence if we already
reply to the method call don't accidentally return positive.
Follow-up for:
d04c1a1c8e7c95daa483d8d52d5fc4c25fbc67f2
Lennart Poettering [Tue, 16 Jan 2024 12:27:26 +0000 (13:27 +0100)]
bus-polkit: explicitly cast away return value we ignore
Lennart Poettering [Tue, 16 Jan 2024 11:53:39 +0000 (12:53 +0100)]
bus-polkit: treat various well-known PK errors as denied
Various recognizable errors from
https://www.freedesktop.org/software/polkit/docs/latest/eggdbus-interface-org.freedesktop.PolicyKit1.Authority.html#eggdbus-errordomain-org.freedesktop.PolicyKit1.Error.
should be considered access failures, hence treat them like that.
Lennart Poettering [Tue, 16 Jan 2024 10:01:28 +0000 (11:01 +0100)]
varlink: also honour new env var $SYSTEMD_VARLINK_LISTEN in varlink_server_listen_auto()
varlink_server_listen_auto() is supposed to be the one-stop solution for
turning simple command line tools into IPC services. They aren't easy to
test/debug however, since you have to invoke them through a service
manager.
Let's make this easier: if the SYSTEMD_VARLINK_LISTEN env var is set,
let's listen on the socket specified therein. This makes things easier
to gdb: just run the service from the cmdline.
Lennart Poettering [Tue, 16 Jan 2024 10:00:02 +0000 (11:00 +0100)]
varlink-idl: prefix all varlink IDL debug lines wit "Varlink-IDL:"
Let's improve debuggability of varlink IPC transactions, and separate
out log messages triggered by the IPC info from others, by prefixing
them with a specific prefix.
Luca Boccassi [Tue, 16 Jan 2024 12:39:46 +0000 (12:39 +0000)]
Merge pull request #30953 from yuwata/network-nexthop-silently-removed-by-kernel
network/nexthop: forget nexthops silently removed by kernel
Yu Watanabe [Sat, 13 Jan 2024 06:39:06 +0000 (15:39 +0900)]
network/route: introduce route_get_link() and use it in log_route_debug()
No functional change, just refactoring and preparation for later
commits.
Sam Leonard [Mon, 15 Jan 2024 17:02:35 +0000 (17:02 +0000)]
vmspawn: supply a serial TTY to the kernel by default
Sam Leonard [Mon, 15 Jan 2024 16:40:09 +0000 (16:40 +0000)]
vmspawn: ignore firmwares with enrolled-keys by default
我超厉害 [Tue, 16 Jan 2024 05:57:07 +0000 (13:57 +0800)]
udev: even if a device is a zac device, scsi-$ID_SERIAL will be reserved for it (#30459)
Co-authored-by: wangyuhang <wangyuhang27@huawei.com>
Yu Watanabe [Tue, 16 Jan 2024 05:55:35 +0000 (14:55 +0900)]
Merge pull request #30851 from lnussel/kernel-install
kernel-install fixes
Daan De Meyer [Mon, 15 Jan 2024 12:46:49 +0000 (13:46 +0100)]
tree-wide: Load entry-token and layout.conf from /usr/lib/kernel/ as well
Yu Watanabe [Mon, 15 Jan 2024 10:48:48 +0000 (19:48 +0900)]
test-network: add test case for removal of nexthop that is a member of a group nexthop
Yu Watanabe [Mon, 15 Jan 2024 04:14:46 +0000 (13:14 +0900)]
network/nexthop: drop dependent nexthops on removal
If a nexthop is removed, dependent nexthops are silently removed by the
kernel. Hence, networkd may be confused that nexthops that depends on the
nexthop still exist, and may fail to configure other routes or so.
Yu Watanabe [Mon, 15 Jan 2024 03:39:19 +0000 (12:39 +0900)]
network/nexthop: introduce a reverse map of nexthop group members
It is not used in this commit, but will be used later.
Preparation for later commits.
Yu Watanabe [Mon, 15 Jan 2024 03:35:47 +0000 (12:35 +0900)]
network/nexthop: read netlink message in nexthop_update_group()
No functional change, preparation for later commits.
Yu Watanabe [Mon, 15 Jan 2024 02:59:21 +0000 (11:59 +0900)]
network/nexthop: reorder elements in NextHop
Daan De Meyer [Mon, 15 Jan 2024 16:16:10 +0000 (17:16 +0100)]
bus-socket: Clarify that inotify is supposed to watch all components
The previous wording of the components could mean that we should only
watch directories, not the socket itself. Reword so that we clearly
mention that all components of the path are watched, including the
socket itself.
Ludwig Nussel [Tue, 9 Jan 2024 11:49:09 +0000 (12:49 +0100)]
kernel-install: silence num kernels installed
Ludwig Nussel [Tue, 9 Jan 2024 11:29:36 +0000 (12:29 +0100)]
kernel-install: fix context_copy
Don't reopen or dup values that weren't set before. Fixes add-all.
Ludwig Nussel [Tue, 9 Jan 2024 16:31:01 +0000 (17:31 +0100)]
strv: introduce strv_copy_unless_empty()
Luca Boccassi [Mon, 15 Jan 2024 14:41:58 +0000 (14:41 +0000)]
Merge pull request #30934 from yuwata/network-introduce-route-nexthop
network/route: introduce RouteNextHop object
Antonio Alvarez Feijoo [Mon, 15 Jan 2024 09:55:31 +0000 (10:55 +0100)]
systemctl-is-system-running: display "offline" with --image
With the `--image` option, the `running_in_chroot` check is not enough. E.g.:
```
> build/systemctl --image /tmp/
20240108 -openSUSE.raw is-system-running
running
```
Yu Watanabe [Mon, 15 Jan 2024 05:19:03 +0000 (14:19 +0900)]
test-network: add simple test case for DHCP relay on bridge interface
For issue #30763.
Yu Watanabe [Sun, 14 Jan 2024 00:40:27 +0000 (09:40 +0900)]
resolve: on_transaction_stream_error() may free multiple transactions
Fixes #30928.
Luca Boccassi [Mon, 15 Jan 2024 10:15:12 +0000 (10:15 +0000)]
Merge pull request #30933 from YHNdnzj/sleep-trivial-cleanup
Trivial cleanup for systemd-sleep unit/man page
Luca Boccassi [Mon, 15 Jan 2024 10:09:30 +0000 (10:09 +0000)]
Merge pull request #30936 from yuwata/network-automatically-reconfigure-interface-on-failure
network: automatically reconfigure interface on failure
Yu Watanabe [Mon, 15 Jan 2024 06:17:57 +0000 (15:17 +0900)]
test-network: try to flip interface state frequently
Yu Watanabe [Mon, 15 Jan 2024 06:40:54 +0000 (15:40 +0900)]
network/link: automatically reconfigure interface on failure
Closes #29246.
Yu Watanabe [Mon, 15 Jan 2024 06:40:07 +0000 (15:40 +0900)]
network/link: shorten code a bit
Yu Watanabe [Sat, 13 Jan 2024 21:22:03 +0000 (06:22 +0900)]
network/route: do not read RTA_OIF twice
It is also read in route_nexthops_read_netlink_message(), and already
stored in the Route object. Let's use it.
Yu Watanabe [Sat, 13 Jan 2024 02:18:53 +0000 (11:18 +0900)]
network/route-nexthop: store ifindex of the assigned interface in Route.nexthop.ifindex
And use it if set on building netlink message.
Yu Watanabe [Fri, 12 Jan 2024 06:23:17 +0000 (15:23 +0900)]
network/route: also use RouteNextHop for managing gateway
No functional change, just refactoring.
Yu Watanabe [Fri, 12 Jan 2024 04:58:23 +0000 (13:58 +0900)]
network/route-nexthop: introduce struct RouteNextHop and replace MultipathRoute with it
It is mostly equivalent to MultipathRoute. So, no functional change,
just refactoring and preparation for later commits.
Luca Boccassi [Sat, 13 Jan 2024 12:12:25 +0000 (12:12 +0000)]
test: create /run/sshd in TEST-74-AUX-UTILS
12264s [ 4819.948632] sshd[1365]: fatal: Missing privilege separation directory: /run/sshd
12264s [ 4819.952120] testsuite-74.sh[1362]: kex_exchange_identification: read: Connection reset by peer
12264s [ 4819.952120] testsuite-74.sh[1362]: Connection reset by /run/ssh-unix-local/socket port 0
https://autopkgtest.ubuntu.com/results/autopkgtest-jammy-upstream-systemd-ci-systemd-ci/jammy/amd64/s/systemd-upstream/20240113_093341_50fc7@/log.gz
We copy binaries manually so some setups are missing, create the directory
as needed
Mike Yuan [Sat, 6 Jan 2024 15:36:24 +0000 (23:36 +0800)]
man/systemd-sleep.conf: document the operation of s2h more thoroughly
Mike Yuan [Sun, 14 Jan 2024 18:58:52 +0000 (02:58 +0800)]
units: update Description= for systemd-sleep units
Mike Yuan [Sat, 6 Jan 2024 14:45:43 +0000 (22:45 +0800)]
battery-util: raise log level for battery_is_discharging_and_low
Luca Boccassi [Sun, 14 Jan 2024 13:03:45 +0000 (13:03 +0000)]
Merge pull request #30932 from yuwata/network-route-split-out-more
network/route: split out more functions to networkd-route-nexthop.[ch]
Cristian Rodríguez [Sat, 13 Jan 2024 23:14:05 +0000 (20:14 -0300)]
Fix gcc14 -Wcalloc-transposed-args warnings
all functions annotated with two parameter _alloc_ are calloc-like.
gcc14 enforces this and warns if arguments are backwards.
Luca Boccassi [Sun, 14 Jan 2024 12:56:42 +0000 (12:56 +0000)]
Merge pull request #30927 from YHNdnzj/logind-action-job
logind-dbus: refuse multiple jobs in method_do_shutdown_or_sleep too
Mike Yuan [Sat, 13 Jan 2024 21:14:42 +0000 (05:14 +0800)]
logind-dbus: refuse multiple jobs in method_do_shutdown_or_sleep too
Fixes #30917
Mike Yuan [Sat, 13 Jan 2024 21:14:25 +0000 (05:14 +0800)]
logind-dbus: minor modernization
Yu Watanabe [Sat, 13 Jan 2024 03:05:32 +0000 (12:05 +0900)]
network/route-nexthop: make multipath_route_get_link() take fallback link
No functional change, just refactoring.
Yu Watanabe [Fri, 12 Jan 2024 04:22:05 +0000 (13:22 +0900)]
network/route-nexthop: do not update MultipathRoute object
A MultipathRoute object is always owned by a Network object, and the
interface matching with its name specified in a MultipathRoute= setting
may be removed, and re-added later with a different ifindex.
Yu Watanabe [Fri, 12 Jan 2024 02:59:09 +0000 (11:59 +0900)]
network/route-nexthop: split out route_nexthops_is_ready_to_configure()
No effective functionality changed, just refactoring and preparation for
later commits.
Yu Watanabe [Fri, 12 Jan 2024 02:07:46 +0000 (11:07 +0900)]
network/route-nexthop: split out route_nexthops_to_string()
And reorder elements shown in the debugging log.
No effective functionality changed, just refactoring.
Yu Watanabe [Tue, 9 Jan 2024 06:04:04 +0000 (15:04 +0900)]
network/route-nexthop: split out netlink message handling for route nexthops
No functional change, just refactoring and preparation for later
commits.
Yu Watanabe [Sat, 13 Jan 2024 11:06:59 +0000 (20:06 +0900)]
varlink/network: OnlineState may not be set
Follow-up for
c5c74d85d302d95d2b0f2b938c4f178e428e19a3 .
Fixes #30918.
Yu Watanabe [Sat, 13 Jan 2024 19:57:05 +0000 (04:57 +0900)]
Merge pull request #30916 from yuwata/network-route-section-verify
network: several cleanups for verifying [Route] section
Yu Watanabe [Tue, 9 Jan 2024 02:45:37 +0000 (11:45 +0900)]
wireguard: verify routes configured in .netdev file
Then, scope and friends are correctly adjusted, and the behavior should
be matched when the same route is configured in .network file.
Yu Watanabe [Sat, 13 Jan 2024 01:06:21 +0000 (10:06 +0900)]
network/route: split out route_section_verify_nexthops()
No effective functional change, just refactoring and preparation for
later commits.
Yu Watanabe [Sat, 13 Jan 2024 01:05:29 +0000 (10:05 +0900)]
network/route: relocate route_type_is_reject() and friends
No functional change, preparation for later commits.
Yu Watanabe [Sat, 13 Jan 2024 00:03:07 +0000 (09:03 +0900)]
Merge pull request #30895 from yuwata/network-drop-object-even-on-fail
network: remove Address object even when we failed to remove relevant address
Luca Boccassi [Fri, 12 Jan 2024 23:43:57 +0000 (23:43 +0000)]
Merge pull request #30902 from YHNdnzj/run-accounting
run: small improvements for accounting data output
Luca Boccassi [Fri, 12 Jan 2024 23:32:12 +0000 (23:32 +0000)]
Merge pull request #30897 from yuwata/network-route-expiration
network: several cleanups for route expiration handling
Luca Boccassi [Fri, 12 Jan 2024 23:27:39 +0000 (23:27 +0000)]
Merge pull request #30896 from yuwata/network-route-conf-parser
network: minor cleanups for conf parsers
Luca Boccassi [Fri, 12 Jan 2024 21:44:51 +0000 (21:44 +0000)]
modprobe: set 'ifb numifbs=0' to avoid autocreating ifb0
Fixes https://github.com/systemd/systemd/issues/30913
Nick Rosbrook [Fri, 12 Jan 2024 19:02:17 +0000 (14:02 -0500)]
test: skip TEST-43-PRIVATEUSER-UNPRIV if unprivileged userns is restricted
With newer versions of AppArmor, unprivileged user namespace creation
may be restricted by default, in which case user manager instances will
not be able to apply PrivateUsers=yes (or the settings which require it).
This can be tested with the kernel.apparmor_restrict_unprivileged_userns
sysctl.
Yu Watanabe [Fri, 12 Jan 2024 21:54:32 +0000 (06:54 +0900)]
Merge pull request #30908 from poettering/nsid
networkd: expose netns "nsid" together with inode id
Frantisek Sumsal [Fri, 12 Jan 2024 18:43:28 +0000 (19:43 +0100)]
network: fix 6rd tunnel link section name
Spotted randomly when going through CI logs:
systemd-udevd[658]: /usr/lib/systemd/network/80-6rd-tunnel.link:21: Unknown section 'Network'. Ignoring.
Follow-up for
658169e6d30 .
Lennart Poettering [Fri, 12 Jan 2024 10:47:45 +0000 (11:47 +0100)]
networkd: expose the nsid in the GetNamespaceId() varlink call
Let's return both ids in the GetNamespaceID(), since they are pretty
much the same concept.
Lennart Poettering [Fri, 12 Jan 2024 10:49:12 +0000 (11:49 +0100)]
networkd: expose nsid via dbus
Lennart Poettering [Fri, 12 Jan 2024 10:49:35 +0000 (11:49 +0100)]
socket-util: add netns_get_nsid() helper to show nsid of netns
Lennart Poettering [Fri, 12 Jan 2024 10:36:49 +0000 (11:36 +0100)]
sd-netlink: hook up nsid APIs
Lennart Poettering [Fri, 12 Jan 2024 17:32:40 +0000 (18:32 +0100)]
Merge pull request #30911 from poettering/vmspawn-tweaks
vmspawn: various clean-ups
chenjiayi [Mon, 8 Jan 2024 11:45:18 +0000 (19:45 +0800)]
udev: distinguish host-managed zoned block in scsi_id
According to SPC4, the value of 0x14 is reserved to distinguish
host managed zoned block, e.g., some SMR (Shingled Magnetic Recording)
disks.
Other utilities such as sg3_utils can successfully recognize such
kind of disks. This patch implements the same ability.
Luca Boccassi [Fri, 12 Jan 2024 15:51:54 +0000 (15:51 +0000)]
Merge pull request #30904 from YHNdnzj/no-selinux-reload
core/execute: don't reload selinux before spawning executor
Daan De Meyer [Thu, 7 Dec 2023 18:52:41 +0000 (19:52 +0100)]
mkosi: Build a directory image by default
Both building and booting a directory image is much faster than
building or booting a disk image so let's default to a directory
image.
In CI, we stick to a disk image to make sure that keeps working as
well.
The only extra dependency this introduces is virtiofsd which is
packaged in all distributions except Debian stable. For users
hacking on systemd on Debian stable, a disk image can be built by
writing the following to mkosi.local.conf:
```
[Output]
Format=disk
```
Lennart Poettering [Fri, 12 Jan 2024 15:02:29 +0000 (16:02 +0100)]
vmspawn: suppress unnecessary '-'
Lennart Poettering [Fri, 12 Jan 2024 14:57:13 +0000 (15:57 +0100)]
vmspawn: use SD_EVENT_SIGNAL_PROCMASK flag where possible
Lennart Poettering [Fri, 12 Jan 2024 14:55:05 +0000 (15:55 +0100)]
vmspawn: use our generic vsock CID parser, instead of a howngrown one
Let's also use 'unsigned' as type for the CID everywhere, and
VMADDR_CID_ANY as marker for "invalid CID", as that's what the vsock
APIs natively do.
Lennart Poettering [Fri, 12 Jan 2024 14:37:43 +0000 (15:37 +0100)]
vmspawn: fix empty lines before section titles in --help text
Always put an empty line before section titles (one was missing so far)
Lennart Poettering [Fri, 12 Jan 2024 14:37:15 +0000 (15:37 +0100)]
vmspawn: make sure are fine with ovmf metadata extensions
The JSON ovmf data on Fedora at least has more fields than we expect,
ignore it.
Yu Watanabe [Fri, 12 Jan 2024 01:00:31 +0000 (10:00 +0900)]
network/ndisc: do not try to set too large value for ICMP ratelimting
Follow-up for
6197db53ba3c61de2268eb723a7a9cd4b3f5f87c .
When we set too large value, the kernel just refuse it. So, this does
not change the net behavior.
Prompted by https://github.com/systemd/systemd/pull/30490#discussion_r1449477125.
Lennart Poettering [Fri, 12 Jan 2024 11:03:00 +0000 (12:03 +0100)]
json: downgrade extension log message to LOG_DEBUG
This is not supposed to be something we should complain about loudly
after all.
Follow-up for:
a617fd904789cd3a05cf4cb2f54649e2a1f73d33
Daan De Meyer [Fri, 12 Jan 2024 08:55:40 +0000 (09:55 +0100)]
Add --root= support for list and prepare add-all for --root= support
Let's make sure these follow the rest of kernel-install and always
operate on the given root directory, even if the verb itself can't
support --root= just yet.
Frantisek Sumsal [Fri, 12 Jan 2024 09:27:43 +0000 (10:27 +0100)]
test: use systemd-id128 from the build dir
As distro sd-id128 might not have all the options we need.
Follow-up for:
378712c
Replaces: #30901
Lennart Poettering [Fri, 12 Jan 2024 09:03:55 +0000 (10:03 +0100)]
man: fix typo AV_VSOCK → AF_VSOCK
A reported by Arian van Putten:
https://github.com/systemd/systemd/pull/30777#pullrequestreview-
1816817988
Mike Yuan [Fri, 12 Jan 2024 09:35:21 +0000 (17:35 +0800)]
run: don't show IP/IO accounting data if 0