Compare commits

...

13 Commits

Author SHA1 Message Date
Vishal Chillara dc81256b26
Merge e9cfc2bd94 into fed7857672 2024-11-07 02:50:04 +01:00
Yu Watanabe fed7857672 NEWS: fix typo
Follow-up for a6d7cc74d6.
2024-11-07 10:05:32 +09:00
Lennart Poettering c8d45ebfd6 update TODO 2024-11-06 22:19:01 +01:00
Lennart Poettering acc8bae0b3 NEWS: various cleanups 2024-11-06 22:18:55 +01:00
Lennart Poettering a6d7cc74d6 NEWS: various cleanups 2024-11-06 21:50:56 +01:00
Luca Boccassi d80d7a2f2a
docs: fix sbsign manpage syntax and add to list, update release instructions (#35055)
<!-- devel-freezer =
{"comment-id":"2460595245","freezing-tag":"v257-rc1"} -->
2024-11-06 20:18:43 +00:00
Luca Boccassi 088793239e docs: add reminder to run update-man-rules before tagging a release 2024-11-06 19:21:14 +00:00
Luca Boccassi 94a46c20da docs: remove 'v' prefix from meson.version
It is actually v-less
2024-11-06 19:20:00 +00:00
Luca Boccassi d6f4c96b10 man: run update-man-rules 2024-11-06 19:19:13 +00:00
Luca Boccassi 9e51b12e13 man: fix syntax error in systemd-sbsign.xml
Follow-up for 5f163921e9
2024-11-06 19:18:35 +00:00
Frantisek Sumsal e9cfc2bd94 varlinkctl: flush stdout after each record in --more mode
So things work correctly even if varlinkctl's output is redirected to a
file.
2024-10-29 22:44:13 +05:30
Vishal Chillara Srinivas 08860aa147 test: resolve: add testing for mDNS browse
Co-authored-by: Frantisek Sumsal <frantisek@sumsal.cz>
Co-authored-by: Vishwanath Chandapur <vishwanath.chandapur@philips.com>
2024-10-29 22:43:02 +05:30
Vishal Chillara Srinivas 2e5bf2774e resolved: Implement continuous mDNS querying as per RFC6762 5.2
Allow for mDNS service/domain/types browsing.
A client can connect to the backend via varlink and receive updates as the
requested service becomes available.
The interval between the first two queries MUST be at least one second,
the intervals between successive queries MUST increase by at least a factor of two.
When the interval between queries reaches or exceeds 60 minutes, a querier MAY cap
the interval to a maximum of 60 minutes, and perform subsequent queries at a
steady-state rate of one query per hour.
Delete expired cache entries one second after goodbye packet received
as per RFC6762 Section 10.1

Cache maintenance:
The querier should plan to issue a query at 80% of the record lifetime, and
then if no answer is received, at 85%, 90%, and 95%.
If an answer is received, then the remaining TTL is reset to the value given
in the answer, and this process repeats for as long as the Multicast DNS querier
has an ongoing interest in the record.
If no answer is received after four queries, the record is deleted when it
reaches 100% of its lifetime.

Co-authored-by: Vishwanath Chandapur <vishwanath.chandapur@philips.com>
2024-10-29 22:38:48 +05:30
28 changed files with 1463 additions and 201 deletions

322
NEWS
View File

@ -103,37 +103,37 @@ CHANGES WITH 257 in spe:
libsystemd:
* systemd's JSON API is now available as public interface of libsystemd
under the name "sd-json". The purpose of the library is to allow
structures to be conveniently created in C code and serialized to
JSON, and for JSON to be conveniently deserialized into in-memory
structures, using callbacks to handle specific keys. Various data
types like integers, floats, booleans, strings, UUIDs, base64-encoded
and hex-encoded binary data, and arrays are supported natively. The
library has been part of systemd for a while as internal component,
and now being made publicly available, too. On major user of sd-json
is the JSON interface sd-varlink (see below). Note that documentation
on sd-json is very much incomplete for now, but the systemd codebase
should provide plenty code real-life code examples.
* systemd's JSON API is now available as public interface of
libsystemd, under the name "sd-json". The purpose of the library is
to allow structures to be conveniently created in C code and
serialized to JSON, and for JSON to be conveniently deserialized into
in-memory structures, using callbacks to handle specific
keys. Various data types like integers, floats, booleans, strings,
UUIDs, base64-encoded and hex-encoded binary data, and arrays are
supported natively. The library has been part of systemd for a while
as internal component, and is now made publicly available. One major
user of sd-json is sd-varlink (see below). Note that the
documentation of sd-json is very much incomplete for now, but the
systemd codebase provides plenty real-life code examples.
* libsystemd's Varlink IPC API is now available as part of libsystemd
* systemd's Varlink IPC API is now available as part of libsystemd,
under the name "sd-varlink". This library is a C implementation of
the Varlink IPC system (https://varlink.org/) that has been adopted
by systemd for various interfaces. It relies on the sd-json JSON
component, see above. Note that documentation on sd-varlink is very
much incomplete for now, but the systemd codebase should provide
plenty code real-life code examples.
component, see above. Note that the documentation of sd-varlink is
very much incomplete for now, but the systemd codebase provides
plenty real-life code examples.
* sd-bus gained a new call sd_bus_pending_method_calls() which returns
the number of currently open asynchronous method calls initiated on
this connection towards peers.
* sd-device gained a new call sd_device_monitor_is_running() that
returns whener the specified monitor object is already running. It
returns whether the specified monitor object is already running. It
also gained sd_device_monitor_get_fd(),
sd_device_monitor_get_events(), sd_device_monitor_get_timeout() and
sd_device_monitor_receive() to permit sd-device to run on a foreign
event loop implementation. It also gained
sd_device_monitor_receive() to permit sd-device to run on top of a
foreign event loop implementation. It also gained
sd_device_get_driver_subsystem() which returns the subsystem of
driver objects. The new sd_device_get_device_id() call returns a
short string identifying the device record.
@ -148,8 +148,9 @@ CHANGES WITH 257 in spe:
* Multipath TCP (MPTCP) is now supported as a socket protocol for
.socket units.
* New /etc/fstab option x-systemd.wants= creates "Wants" dependencies.
(This is similar to the previously available x-systemd.requires=.)
* A new /etc/fstab option x-systemd.wants= creates "Wants="
dependencies. (This is similar to the previously available
x-systemd.requires=.)
* The initialization of the system clock during boot and updates has
been simplified: both PID 1 or systemd-timesyncd will pick the latest
@ -161,17 +162,17 @@ CHANGES WITH 257 in spe:
shutdown, so that the user may use it to initiate a reboot if the
system freezes otherwise.
* The new unit option PrivateUsers=identity can be used to request a
user namespace with an identity mapping for the first 65536
UIDs/GIDs. This is analogous to the systemd-nspawn's
* The new value "identity" for the unit setting PrivateUsers= may be
used to request a user namespace with an identity mapping for the
first 65536 UIDs/GIDs. This is analogous to the systemd-nspawn's
--private-users=identity.
* The new unit option PrivateTmp=disconnected can be used to specify
that a separate tmpfs instance should be used for /tmp/ and /var/tmp/
for the unit.
* The new value "disconnected" for the unit setting PrivateTmp= may be
used to specify that a separate tmpfs instance should be used for
/tmp/ and /var/tmp/ for the unit.
* The manager (and various other tools too) use pidfds in more places
to refer to processes.
* The server manager (and various other tools too) use pidfds in more
places to refer to processes.
* A build option -D link-executor-shared=false can be used to build
the systemd-executor binary (added in a previous release) in a way
@ -185,41 +186,41 @@ CHANGES WITH 257 in spe:
execute.
* The systemd.machine_id= kernel command line parameter interpreted by
PID 1 now supports an additional special value: if "firmware" is
specified the machine ID is initialized from the SMBIOS/DeviceTree
system UUID. (Previously this was already done in VM environments,
this extends the concept to any system, but only on explicit request
via this option.)
PID 1 now supports an additional special value: if set to "firmware"
the machine ID is initialized from the SMBIOS/DeviceTree system
UUID. (Previously this was already done automatically in VM
environments, this extends the concept to any system, but only on
explicit request via this option.)
* The ImportCredential= setting in service unit files now permits
renaming credentials imported.
renaming of credentials as they are imported.
* The RestartMode= gained a new "debug" setting. If specified and the
service fails so that it shall be restarted it is invoked in
* The RestartMode= setting gained a new "debug" value. If specified and
the service fails so that it shall be restarted it is invoked in
"debugging mode". Debugging mode means that the $DEBUG_INVOCATION
environment variable will be set to "1" for the new
invocation. Moreover, any setting LogLevelMax= will be temporarily
changed to "debug" for the next invocation. This mode is useful to
repeat invocation of tools if they fail but with additional logging
or testing routines turned on.
automatically repeat invocation of tools in case they fail but with
additional logging or testing routines enabled.
* A new service setting BindLogSockets= has been added that
controls whether the AF_UNIX sockets required for logging shall be
bind mounted to the mount sandbox allocated for the service.
* PID 1 will now optionally load a policy for the new Linux IPE LSM at
boot.
* At early boot, PID 1 will now optionally load a policy for the new
Linux IPE LSM.
* Transient services (StartTransientUnit() D-Bus method) may now
receive additional, arbitrary file descriptors to pass to executed
service processes on activation using the new ExtraFileDescriptor=
unit property.
* Transient services (as invoked by the StartTransientUnit() D-Bus
method) may now receive additional, arbitrary file descriptors to
pass to executed service processes during activation using the new
ExtraFileDescriptor= unit property.
* Calendar .timer units gained a new boolean DeferReactivation=
option. If enabled and the repetitive calendar timer elapses again
while the service the timer activates is still running, immediate
reactivation once it finishes is skipped, and the timer has to elapse
again before the service is reactivated.
reactivation of the service once it finishes is skipped, and the
timer has to elapse again before the service is reactivated.
* Generator processes invoked by the service manager will now receive a
new environment variable $SYSTEMD_SOFT_REBOOTS_COUNT that indicates
@ -245,10 +246,10 @@ CHANGES WITH 257 in spe:
"strict" a new cgroup namespace is allocated for the service, and
cgroupfs is mounted read-only for the service.
* The StateDirectory=, RuntimeDirectory=, CacheDirectory=, LogsDirectory=,
and ConfigurationDirectory= settings gained support for configuring the
respective directories as read-only, via a ':ro' flag that can be
appended to each setting.
* The StateDirectory=, RuntimeDirectory=, CacheDirectory=,
LogsDirectory=, and ConfigurationDirectory= settings gained support
for configuring the respective directories as read-only, via a ':ro'
flag that can be appended to each setting's value.
* When DynamicUser= is combined with
StateDirectory=/RuntimeDirectory=/CacheDirectory=/LogsDirectory= and
@ -258,15 +259,15 @@ CHANGES WITH 257 in spe:
chown()ing.
* A new service property PrivatePIDs= has been added that runs executed
processes as PID 1 - the init process - within their own PID namespace.
PrivatePIDs= also mounts /proc/ so only processes within the new PID
namespace are visible.
processes as PID 1 - the init process - within their own PID
namespace. PrivatePIDs= also mounts /proc/ so only processes within
the new PID namespace are visible.
systemd-udevd:
* udev rules now set 'uaccess' for /dev/udmabuf, giving locally
logged-in users access to the hardware. This is necessary to support
IPMI cameras with libcamera.
logged-in users access to the hardware. This is useful in order to
support IPMI cameras with libcamera.
* Serial port devices will no longer show up as systemd units, unless
they have an IO port or memory assigned to them. This means that only
@ -281,9 +282,9 @@ CHANGES WITH 257 in spe:
searched for both on the interface's parent device (as before) and
the device itself (new).
* Various USB hardware wallets have are now recognized by udev via a
.hwdb file, and get the ID_HARDWARE_WALLET= property set, which
enables "uaccess" for them, i.e. direct unprivileged access.
* Various USB hardware wallets are now recognized by udev via a .hwdb
file, and get the ID_HARDWARE_WALLET= property set, which enables
"uaccess" for them, i.e. direct unprivileged access.
* udevadm info will now output the device ID string in lines prefixed
with "J:", and the driver subsystem in lines prefixed with "B:".
@ -293,8 +294,8 @@ CHANGES WITH 257 in spe:
systemd-logind:
* New DesignatedMaintenanceTime= configuration option allows
shutdowns to be automatically scheduled at the specified time.
* New DesignatedMaintenanceTime= configuration option allows shutdowns
to be automatically scheduled at the specified time.
* logind now reacts to Ctrl-Alt-Shift-Esc being pressed. It will send
out a org.freedesktop.login1.SecureAttentionKey signal, indicating a
@ -308,8 +309,8 @@ CHANGES WITH 257 in spe:
session switches away.
* systemd-logind now exposes two D-Bus properties CanLock and CanIdle
for all sessions that indicate whether the session's class supports
screen locking and idle detection.
for all sessions. These properties indicate whether the session's
class supports screen locking and idleness detection.
* systemd-inhibit now allows interactive polkit authorization. It
gained a --no-ask-password option to suppress it.
@ -320,12 +321,13 @@ CHANGES WITH 257 in spe:
Machines started via the systemd-vmspawn@.service unit will now be
registered with systemd-machined.
* systemd-machined gained a pretty complete set of Varlink interfaces
to its functionality as alternative to the existing D-Bus interface.
* systemd-machined gained a pretty complete set of Varlink APIs
exposing its functionality. This is an alternative to the
pre-existing D-Bus interface.
systemd-resolved:
* resolvconf command now supports '-p' switch. If specified, the
* The resolvconf command now supports '-p' switch. If specified, the
interface will not be used as the default route for domain name
lookups.
@ -337,11 +339,11 @@ CHANGES WITH 257 in spe:
* IPv6 address labels can be configured in a new [IPv6AddressLabel]
section with Prefix= and Label= settings.
* 'networkctl edit' can now read the new contents from standard input
with the new --stdin option.
* 'networkctl edit' can now read the new file contents from standard
input with the new --stdin option.
* 'networkctl edit' and 'cat' now supports editing .netdev files by
link. 'networkctl cat' can also list all configuration files
* 'networkctl edit' and 'cat' now support editing/showing .netdev files
by link. 'networkctl cat' can also list all configuration files
associated with an interface at once with ':all'.
* networkctl gained a --no-ask-password option to suppress interactive
@ -350,7 +352,7 @@ CHANGES WITH 257 in spe:
* "mac" has been added to the default AlternativeNamesPolicy= setting
for network links (via 99-default.link). This means "enx*" interface
names will now be added to the list of alternative interface names by
default for all interfaces that have a MAC address assigned to them
default, for all interfaces that have a MAC address assigned
by hardware.
* networkd .netdev bridge devices gained a new setting FDBMaxLearned=
@ -365,18 +367,18 @@ CHANGES WITH 257 in spe:
thus highlighting conflict of ownership/management of these knobs.
* systemd-networkd will now make RFC9463 DNR fields available to
systemd-resolved, for automatic DoT configuration, and similar.
systemd-resolved, for automatic DNS DoT configuration, and similar.
systemd-boot, systemd-stub, and related tools:
* The EFI stub now supports loading of .ucode sections with microcode
from PE add-on files. It now also supports loading .initrd sections
from PE add-on files. It also now supports loading .initrd sections
from PE add-on files.
* A new .profile PE section type is now documented and supported in
systemd-measure, ukify, systemd-stub and systemd-boot. Those new
systemd-measure, ukify, systemd-stub and systemd-boot. These new
sections allow multiple "profiles" to be stored together in the UKI,
with .profile sections creating groupings of sections in the UKI,
where each .profile section creates groupings of sections in the UKI,
allowing some sections to be shared and other sections like .cmdline
or .initrd unique to the profile. This may be used to provide a
single UKI that synthesizes multiple menu items in the boot menu (for
@ -389,10 +391,10 @@ CHANGES WITH 257 in spe:
can contain multiple .dtbauto sections, and the 'compatible' string
therein will be compared with the equivalent field in the DTB
provided by the firmware, if present. If absent, SMBIOS will be used
to calculate hardware IDs and compare them with the content of
.hwids. This allows including multiple DTBs in a single UKI, with
the bootloader automatically selecting the correct one for the
current hardware.
to calculate hardware IDs (CHIDs) and look them up in the content of
.hwids, hopefully revealing an fallback 'compatible' string. This
allows including multiple DTBs in a single UKI, with systemd-stub
automatically loading the correct one for the current hardware.
* ukify gained an --extend switch to import an existing UKI to
be extended, and a --measure-base= switch to support measurement
@ -405,25 +407,26 @@ CHANGES WITH 257 in spe:
* systemd-stub will report the partition UUID and image identifier its
UKI executable is placed on separately from the data systemd-boot
provides about where to find its own executable. This is useful when
systemd-boot and UKIs are placed on distinct partitions (i.e. ESP and
XBOOTLDR).
provides about where to find its own executable, via EFI
variables. This is useful when systemd-boot and UKIs are placed on
distinct partitions (i.e. ESP and XBOOTLDR).
* bootctl --print-loader-path and --print-stub-path that output the
path to the boot loader or UKI used for the current boot.
* bootctl gained new switches --print-loader-path and --print-stub-path
that output the path to the boot loader or UKI used for the current
boot.
* bootctl kernel-identify now supports identifying EFI add-ons.
* bootctl kernel-identify now recognizes EFI add-ons.
* bootctl gained a --random-seed=yes|no option to control provisioning
of the random seed file in ESP. (This is useful when producing an
image that will be used multiple times.)
of the random seed file in the ESP. (This is useful when producing an
image that will be used in multiple instances.)
* bootctl now optionally supports installing UEFI Secure Boot databases
(ESLs) for systemd-boot to pick up and automatically enroll if the
system is booted in Setup Mode. This is controlled via bootctl's new
--secure-boot-auto-enroll=yes switch (and some auxiliary ones). A
certificate can be provided in DER format, and it is automatically
converted into an ESL, as needed.
(i.e. db/dbx/… databases in ESL format) for systemd-boot to pick up
and automatically enroll if the system is booted in Setup Mode. This
is controlled via bootctl's new --secure-boot-auto-enroll=yes switch
(and some auxiliary ones). A certificate can be provided in DER
format, and is automatically converted into an ESL, as needed.
* bootctl, systemd-measure, systemd-repart when referencing signing
keys on OpenSSL engines may now query for PINs and similar via
@ -431,9 +434,9 @@ CHANGES WITH 257 in spe:
caching and UI).
* A new systemd-sbsign tool has been added, that can be used to sign
EFI binaries (PE). This tool supports OpenSSL engines and providers,
with pin caching support for PKCS11. ukify supports it as an
alternative to sbsigntool and pesign.
EFI binaries (PE) for Secure Boot. This tool supports OpenSSL engines
and providers, with pin caching support for PKCS11. ukify supports it
as an alternative to sbsigntool and pesign.
The journal:
@ -468,22 +471,22 @@ CHANGES WITH 257 in spe:
and AppStream metadata.
* Transfer definitions for systemd-sysupdate are supposed to carry the
".transfer" suffix now, changing from ".conf". The latter is
supported for compatibility too, but it's recommended to rename all
files reflecting this suffix change.
".transfer" suffix now, changing from ".conf". The latter remains
supported for compatibility, but it's recommended to rename all files
reflecting this suffix change.
* systemd-sysupdate now supports a new ".feature" files that may be
* systemd-sysupdate now supports new ".feature" files that may be
used in conjunction with ".transfer" files to group them together, and
allow them to be turned off or on, individually per group.
TPM & systemd-cryptsetup:
* The 'tpm2' verb which lists usable TPM2 devices has been moved from
systemd-creds to systemd-analyze.
* The 'has-tpm2' verb which reports whether TPM2 functionality is
available has been moved from systemd-creds to systemd-analyze.
* systemd-tpm2-setup will gracefully handle TPMs that have a PIN set on
the TPM, and not automatically set up a Storage Root Key (SRK) in
that case.
the TPM, and not attempt to automatically set up a Storage Root Key
(SRK) in that case.
* New crypttab option password-cache=yes|no|read-only can be used to
customize password caching.
@ -525,7 +528,7 @@ CHANGES WITH 257 in spe:
start the specified executable on the remote side, and communicate
with the remote process using the Varlink protocol.
"ssh:" address specification has been renamed to "ssh-unix:"
The "ssh:" address specification has been renamed to "ssh-unix:"
(reflecting the fact it is used to connect to a remote AF_UNIX socket
via SSH). The old syntax is still supported for backwards
compatibility.
@ -546,7 +549,8 @@ CHANGES WITH 257 in spe:
to enable internal compression in filesystems created offline.
* systemd-repart understands a new MakeSymlinks= option to create one
or more symlinks (each specified as a symlink name and target).
or more symlinks (each specified as a symlink name and target) within
a newly formatted file system.
* systemd-repart gained a new SupplementFor= setting that allows
allocating a partition only if some other existing partition cannot
@ -559,15 +563,15 @@ CHANGES WITH 257 in spe:
systemd-ssh-proxy:
* systemd-ssh-proxy now also supports the "VSOCK MUX" protocol used by
CloudHypervisor/Firecracker to expose AF_VSOCK sockets of the VM on
the host. Or in other words: it's now possible to directly connect to
ssh via AF_VSOCK from hosts to VMs of these two hypervisors
(previously this was only supported for hypervisors which expose
AF_VSOCK on the host as AF_VSOCK, such as qemu).
* systemd-ssh-proxy now also supports the AF_UNIX-based "VSOCK MUX"
protocol used by CloudHypervisor/Firecracker to expose AF_VSOCK
sockets of the VM on the host. Or in other words: it's now possible
to directly connect to ssh via AF_VSOCK from hosts to VMs of these
two hypervisors (previously this was only supported for hypervisors
which expose AF_VSOCK on the host as AF_VSOCK, such as qemu).
* systemd-ssh-proxy can now reference local VMs by their name: connect
to any local VM "foobar" registered with machined via "ssh
to any local VM "foobar" registered with systemd-machined via "ssh
machine/foobar" using the AF_VSOCK protocol.
systemd-analyze:
@ -591,7 +595,6 @@ CHANGES WITH 257 in spe:
* 'busctl monitor' gained new options --limit-messages= and --timeout=
to set the number of matches or limit the runtime of the command.
This is intended to be used in scripts.
* busctl now supports doing method calls with embedded unix file
descriptors.
@ -609,9 +612,9 @@ CHANGES WITH 257 in spe:
systemd-importd:
* A new generator sytemd-import-generator has been added to
synthetisize image download jobs. This provides functionality similar
to importctl, but configured via the kernel command line and system
* A new generator sytemd-import-generator has been added to synthesize
image download jobs. This provides functionality similar to
importctl, but is configured via the kernel command line and system
credentials. It may be used to automatically download sysext,
confext, portable service, nspawn container or vmspawn VM images at
boot.
@ -640,6 +643,32 @@ CHANGES WITH 257 in spe:
systemd-homed to allow users to change selected properties of their
own user records.
systemd-run & run0:
* run0 gained a new pair of settings --pty and --pipe that control
whether to invoke the specified binary on a freshly allocated pseudo
TTY, or whether to pass the client's STDIN/STDOUT/STDERR through
directly.
* run0 gained a new switch --shell-prompt-prefix= that permits passing
in a string to display on each shell prompt as prefix. If not
specified otherwise this will show a superhero emoji (🦸), in order
to visually communicate the temporarily elevated privileges a run0
session provides. This makes use of the $SHELL_PROMPT_PREFIX
environment variables mentioned below.
* systemd-run can output some of its runtime data in JSON format via
the new --json= option.
systemd-tmpfiles:
* systemd-tmpfiles --purge switch now requires specification of at
least one tmpfiles.d/ drop-in file.
* tmpfiles.d/ files gained a new '?' specifier for the 'L' line type to
create a symlink only if the source exists, and gracefully skip the
line otherwise.
Miscellaneous:
* systemctl now supports the --now option with the 'reenable' verb.
@ -654,21 +683,13 @@ CHANGES WITH 257 in spe:
* localectl gained a -l/--full option to show output without
ellipsization.
* systemd-run can output some data as JSON via the new --json= option.
* timedatectl now supports interactive polkit authorization.
* systemd-tmpfiles --purge switch now requires specification of at
least one tmpfiles.d/ drop-in file.
* tmpfiles.d gained a new '?' specifier for the 'L' type to create a
symlink only if the source exists, and gracefully skip otherwise.
* The new Linux mseal(), listmount(), statmount() syscalls have been
added to relevant system call groups.
* The systemd-ask-password concept has been extended with a per-user
concept, i.e. user programs may now ask for passwords via the same
* The systemd-ask-password logic has been extended with a per-user
scope, i.e. user programs may now ask for passwords via the same
mechanism and the previously system-wide only mechanism.
* A new set of system/service credentials are added:
@ -681,17 +702,8 @@ CHANGES WITH 257 in spe:
useful to visually highlight the fact a specific shell prompt
originates from a specific system, execution context or tool. These
credentials and environment variables are supposed to be generically
useful within and outside of the immediate systemd context.
* run0 gained a new pair of settings --pty and --pipe that control
whether to invoke the specified binary on a freshly allocated pseudo
TTY, or whether to pass the client's STDIN/STDOUT/STDERR through
directly. run0 also gained a new switch --shell-prompt-prefix= that
permits passing in a string to display on each shell prompt as
prefix. If not specified otherwise this will show a superman emoji
(🦸), in order to visually communicate the temporarily elevated
privileges a run0 session provides. This makes use of the
$SHELL_PROMPT_PREFIX environment variables mentioned above.
useful within and outside of the immediate systemd context. It is
also used by 'run0', see above.
* New RELEASE_TYPE=, EXPERIMENT=, EXPERIMENT_URL= fields have been
defined for the /etc/os-release file. For example,
@ -718,28 +730,28 @@ CHANGES WITH 257 in spe:
https://github.com/microsoft/terminal/pull/8055
https://conemu.github.io/en/AnsiEscapeCodes.html#ConEmu_specific_OSC
* systemd-sysusers is now able to create fully locked accounts. For
compatibility it so far created accounts with a locked (i.e. invalid)
password, but not marked locked as a whole. With the new "!" modifier
for "u" lines, it is now possible to create fully locked
accounts. The distinction between accounts with a locked password and
fully locked accounts is relevant when considering non-password forms
of authentication, i.e. SSH and such. It is strongly recommended to
make use of this new feature for almost all system accounts, since
they usually do not require (and should not permit) interactive
logins. All of systemd's own system users have been changed to be
marked as fully locked.
* systemd-sysusers is now able to create fully locked user
accounts. For compatibility it so far created accounts with a locked
(i.e. invalid) password, but not marked locked as a whole. With the
new "!" modifier for "u" lines, it is now possible to create fully
locked accounts. The distinction between accounts with a locked
password and fully locked accounts is relevant when considering
non-password forms of authentication, i.e. SSH and such. It is
strongly recommended to make use of this new feature for almost all
system accounts, since they usually do not require (and should not
permit) interactive logins. All of systemd's own system users have
been changed to be marked as fully locked.
* systemd-coredump now supports a new EnterNamespace= option, which
defaults to off. If enabled systemd-coredump will access the mount
namespace of any crashed process to acquire debug symbol information,
in order to be able to symbolized backtraces. This option is useful
to improve backtraces of processes of containerized
applications. (Note that the host systemd-coredump preferably
dispatches coredump processing to the container itself, if it
supports that. Only full-OS containers which run systemd inside will
support this however, in which case EnterNamespace= might be an
alternative approach to acquire symbolized backtraces.)
in order to be able to symbolize backtraces. This option is useful to
improve backtraces of processes of containerized applications. (Note
that the host systemd-coredump preferably dispatches coredump
processing to the container itself, if it supports that. Only full-OS
containers which run systemd inside will support this however, in
other cases EnterNamespace= might be an suitable approach to acquire
symbolized backtraces.)
Contributions from: A. Wilcox, Abderrahim Kitouni, Adrian Vovk,
Alain Greppin, Allison Karlitskaya, Alyssa Ross, Anders Jonsson,

6
TODO
View File

@ -129,6 +129,12 @@ Deprecations and removals:
Features:
* machined: when registering a machine, also take a relative cgroup path,
relative to the machine's unit. This is useful when registering unpriv
machines, as they might sit down the cgroup tree, below a cgroup delegation
boundary. Then, install an inotify watch on that cgroup to track when the
machine's local cgroup goes down.
* resolved: report ttl in resolution replies if we know it. This data is useful
for tools such as wireguard which want to periodically re-resolve DNS names,
and might want to use the TTL has hint for that.

View File

@ -15,18 +15,19 @@ SPDX-License-Identifier: LGPL-2.1-or-later
6. [RC1] Update library numbers in `meson.build`
7. Update version number in `meson.version` (e.g. from `256~devel` to `256~rc1` or from `256~rc3` to `256`). Note that this uses a tilde (\~) instead of a hyphen (-) because tildes sort lower in version comparisons according to the [version format specification](https://uapi-group.org/specifications/specs/version_format_specification/), and we want `255~rc1` to sort lower than `255`.
8. Check dbus docs with `ninja -C build update-dbus-docs`
9. Update translation strings (`ninja -C build systemd-pot`, `ninja -C build systemd-update-po`) - drop the header comments from `systemd.pot` + re-add SPDX before committing. If the only change in a file is the 'POT-Creation-Date' field, then ignore that file.
10. Tag the release: `version="v$(sed 's/~/-/g' meson.version)" && git tag -s "${version}" -m "systemd ${version}"` (tildes are replaced with hyphens, because git doesn't accept the former).
11. Do `ninja -C build`
12. Make sure that the version string and package string match: `build/systemctl --version`
13. [FINAL] Close the github milestone and open a new one (https://github.com/systemd/systemd/milestones)
14. "Draft" a new release on github (https://github.com/systemd/systemd/releases/new), mark "This is a pre-release" if appropriate.
15. Check that announcement to systemd-devel, with a copy&paste from NEWS, was sent. This should happen automatically.
16. Update IRC topic (`/msg chanserv TOPIC #systemd Version NNN released | Online resources https://systemd.io/`)
17. [FINAL] Create an empty -stable branch: `git push systemd origin/main:refs/heads/v${version}-stable`.
18. [FINAL] Build and upload the documentation (on the -stable branch): `ninja -C build doc-sync`
9. Check manpages list with `ninja -C build update-man-rules`
10. Update translation strings (`ninja -C build systemd-pot`, `ninja -C build systemd-update-po`) - drop the header comments from `systemd.pot` + re-add SPDX before committing. If the only change in a file is the 'POT-Creation-Date' field, then ignore that file.
11. Tag the release: `version="v$(sed 's/~/-/g' meson.version)" && git tag -s "${version}" -m "systemd ${version}"` (tildes are replaced with hyphens, because git doesn't accept the former).
12. Do `ninja -C build`
13. Make sure that the version string and package string match: `build/systemctl --version`
14. [FINAL] Close the github milestone and open a new one (https://github.com/systemd/systemd/milestones)
15. "Draft" a new release on github (https://github.com/systemd/systemd/releases/new), mark "This is a pre-release" if appropriate.
16. Check that announcement to systemd-devel, with a copy&paste from NEWS, was sent. This should happen automatically.
17. Update IRC topic (`/msg chanserv TOPIC #systemd Version NNN released | Online resources https://systemd.io/`)
18. [FINAL] Create an empty -stable branch: `git push systemd origin/main:refs/heads/v${version}-stable`.
19. [FINAL] Build and upload the documentation (on the -stable branch): `ninja -C build doc-sync`
20. [FINAL] Change the Github Pages branch to the newly created branch (https://github.com/systemd/systemd/settings/pages) and set the 'Custom domain' to 'systemd.io'
21. [FINAL] Update version number in `meson.version` to the devel version of the next release (e.g. from `v256` to `v257~devel`)
21. [FINAL] Update version number in `meson.version` to the devel version of the next release (e.g. from `256` to `257~devel`)
# Steps to a Successful Stable Release

View File

@ -1069,6 +1069,7 @@ manpages = [
'ENABLE_RFKILL'],
['systemd-run-generator', '8', [], ''],
['systemd-run', '1', [], ''],
['systemd-sbsign', '1', [], ''],
['systemd-sleep.conf', '5', ['sleep.conf.d'], ''],
['systemd-socket-activate', '1', [], ''],
['systemd-socket-proxyd', '8', [], ''],

View File

@ -83,7 +83,7 @@
<varlistentry>
<term><option>--private-key=<replaceable>PATH/URI</replaceable></option></term>
<term><option>--private-key-source=<replaceable>TYPE</replaceable>[:<replaceable>NAME<replaceable>]</option></term>
<term><option>--private-key-source=<replaceable>TYPE</replaceable>[:<replaceable>NAME</replaceable>]</option></term>
<term><option>--certificate=<replaceable>PATH</replaceable></option></term>
<listitem><para>Set the Secure Boot private key and certificate for use with the

View File

@ -8,6 +8,7 @@ basic_dns_sources = files(
'resolved-dns-rr.c',
'resolved-dns-answer.c',
'resolved-dns-question.c',
'resolved-dns-browse-services.c',
'resolved-util.c',
'dns-type.c',
)

View File

@ -0,0 +1,687 @@
/* SPDX-License-Identifier: LGPL-2.1-or-later */
#include "af-list.h"
#include "event-util.h"
#include "json-util.h"
#include "random-util.h"
#include "resolved-dns-browse-services.h"
#include "resolved-dns-cache.h"
#include "resolved-varlink.h"
/* RFC 6762 section 5.2 - The querier should plan to issue a query at 80% of
* the record lifetime, and then if no answer is received, at 85%, 90%, and 95%. */
static usec_t mdns_maintenance_next_time(usec_t until, uint32_t ttl, int ttl_state) {
return usec_sub_unsigned(until, (20 - ttl_state * 5) * ttl * USEC_PER_SEC / 100);
}
/* RFC 6762 section 5.2 - A random variation of 2% of the record TTL should
* be added to maintenance queries. */
static usec_t mdns_maintenance_jitter(uint32_t ttl) {
return random_u64_range(100) * 2 * ttl * USEC_PER_SEC / 10000;
}
#define MDNS_80_PERCENT 80
#define MDNS_5_PERCENT 5
static void mdns_find_service_from_query(DnsService **service, DnsServiceBrowser *sb, DnsQuery *q) {
assert(sb);
/* Find the service that owns the query. */
LIST_FOREACH(dns_services, s, sb->dns_services) {
if (s->query == q) {
*service = s;
return;
}
}
*service = NULL;
}
static void mdns_maintenance_query_complete(DnsQuery *q) {
_cleanup_(dns_service_browser_unrefp) DnsServiceBrowser *sb = NULL;
_cleanup_(dns_query_freep) DnsQuery *query = q;
DnsService *service = NULL;
int r;
assert(query);
assert(query->manager);
sb = dns_service_browser_ref(hashmap_get(query->manager->dns_service_browsers, query->varlink_request));
if (!sb)
return;
if (query->state != DNS_TRANSACTION_SUCCESS) {
r = 0;
goto finish;
}
r = dns_answer_match_key(query->answer, sb->key, NULL);
if (r <= 0)
goto finish;
r = mdns_browser_lookup_cache(sb, query->answer_family);
finish:
if (r < 0)
log_error_errno(r, "mDNS maintenance query complete failed: %m");
mdns_find_service_from_query(&service, sb, query);
if (service)
service->query = NULL;
}
static int mdns_maintenance_query(sd_event_source *s, uint64_t usec, void *userdata) {
DnsService *service = NULL;
_cleanup_(dns_query_freep) DnsQuery *q = NULL;
int r;
assert(userdata);
service = userdata;
if (service->rr_ttl_state++ == MDNS_TTL_100_PERCENT)
return mdns_browser_lookup_cache(service->sb, service->family);
r = dns_query_new(service->sb->m, &q, service->sb->question_utf8, service->sb->question_idna, NULL, service->sb->ifindex, service->sb->flags);
if (r < 0)
goto finish;
q->complete = mdns_maintenance_query_complete;
q->varlink_request = sd_varlink_ref(service->sb->link);
service->query = TAKE_PTR(q);
usec_t next_time = mdns_maintenance_next_time(service->until, service->rr->ttl, service->rr_ttl_state);
/* Schedule next maintenance query for service */
r = event_reset_time(
service->sb->m->event, &service->schedule_event,
CLOCK_BOOTTIME, next_time, 0, mdns_maintenance_query,
service, 0, "mdns-next-query-schedule", true);
if (r < 0)
goto finish;
r = dns_query_go(service->query);
if (r < 0)
goto finish;
return 0;
finish:
dns_query_free(service->query);
return log_error_errno(r, "Failed mdns maintenance query: %m");
}
int dns_add_new_service(DnsServiceBrowser *sb, DnsResourceRecord *rr, int owner_family) {
_cleanup_(dns_service_unrefp) DnsService *s = NULL;
int r;
assert(sb);
assert(rr);
s = new(DnsService, 1);
if (!s)
return log_oom();
usec_t usec = now(CLOCK_BOOTTIME);
*s = (DnsService) {
.n_ref = 1,
.sb = dns_service_browser_ref(sb),
.rr = dns_resource_record_copy(rr),
.family = owner_family,
.until = rr->until,
.query = NULL,
.rr_ttl_state = MDNS_TTL_80_PERCENT,
};
LIST_PREPEND(dns_services, sb->dns_services, s);
/* Schedule the first cache maintenance query at 80% of the record's TTL.
* Subsequent queries issued at 5% increments until 100% of the TTL. RFC 6762 section 5.2.
* If service is being added after 80% of the TTL has already elapsed,
* schedule the next query at the next 5% increment. */
usec_t next_time = 0;
while (s->rr_ttl_state <= MDNS_TTL_100_PERCENT) {
next_time = mdns_maintenance_next_time(rr->until, rr->ttl, s->rr_ttl_state);
if (next_time >= usec)
break;
s->rr_ttl_state++;
}
if (next_time < usec) {
/* If next_time is still in the past, the service is being added after it has already expired.
* Just schedule a 100% maintenance query */
next_time = usec + USEC_PER_SEC;
s->rr_ttl_state = MDNS_TTL_100_PERCENT;
}
usec_t jitter = mdns_maintenance_jitter(rr->ttl);
r = sd_event_add_time(
sb->m->event,
&s->schedule_event,
CLOCK_BOOTTIME,
usec_add(next_time, jitter),
0,
mdns_maintenance_query,
s);
if (r < 0)
return r;
TAKE_PTR(s);
return 0;
}
void dns_remove_service(DnsServiceBrowser *sb, DnsService *service) {
assert(sb);
assert(service);
LIST_REMOVE(dns_services, sb->dns_services, service);
dns_service_free(service);
}
DnsService *dns_service_free(DnsService *service) {
if (!service)
return NULL;
sd_event_source_disable_unref(service->schedule_event);
if (service->query && DNS_TRANSACTION_IS_LIVE(service->query->state))
dns_query_complete(service->query, DNS_TRANSACTION_ABORTED);
dns_service_browser_unref(service->sb);
dns_resource_record_unref(service->rr);
return mfree(service);
}
DEFINE_TRIVIAL_REF_UNREF_FUNC(DnsService, dns_service, dns_service_free);
int mdns_service_update(DnsService *service, DnsResourceRecord *rr, usec_t t) {
service->until = rr->until;
service->rr->ttl = rr->ttl;
/* Update the 80% TTL maintenance event based on new record received from the network.
* RFC 6762 section 5.2 */
usec_t next_time = mdns_maintenance_next_time(service->until, service->rr->ttl, MDNS_TTL_80_PERCENT);
usec_t jitter = mdns_maintenance_jitter(service->rr->ttl);
if (service->schedule_event)
return sd_event_source_set_time(service->schedule_event, usec_add(next_time, jitter));
return 0;
}
bool dns_service_contains(DnsService *services, DnsResourceRecord *rr, int owner_family) {
usec_t t = now(CLOCK_BOOTTIME);
LIST_FOREACH(dns_services, service, services)
if (dns_resource_record_equal(rr, service->rr) > 0 && service->family == owner_family) {
if (rr->ttl <= 1)
return true;
if (rr->until > service->until)
mdns_service_update(service, rr, t);
return true;
}
return false;
}
void dns_browse_services_purge(Manager *m, int family) {
int r = 0;
/* Called after caches are flused.
* Clear local service records and notify varlink client. */
if (!(m && m->dns_service_browsers))
return;
DnsServiceBrowser *sb;
HASHMAP_FOREACH(sb, m->dns_service_browsers) {
r = sd_event_source_set_enabled(sb->schedule_event, SD_EVENT_OFF);
if (r < 0)
goto finish;
if (family == AF_UNSPEC) {
r = mdns_browser_lookup_cache(sb, AF_INET);
if (r < 0)
goto finish;
r = mdns_browser_lookup_cache(sb, AF_INET6);
if (r < 0)
goto finish;
return;
}
r = mdns_browser_lookup_cache(sb, family);
if (r < 0)
goto finish;
}
finish:
if (r < 0)
log_error_errno(r, "mdns browse services purge failed: %m");
return;
}
int mdns_manage_services_answer(DnsServiceBrowser *sb, DnsAnswer *answer, int owner_family) {
DnsResourceRecord *i;
_cleanup_(sd_json_variant_unrefp) sd_json_variant *array = NULL;
int r;
assert(sb);
/* Check for new service added */
DNS_ANSWER_FOREACH(i, answer) {
_cleanup_free_ char *name = NULL, *type = NULL, *domain = NULL;
_cleanup_(sd_json_variant_unrefp) sd_json_variant *entry = NULL;
if (dns_service_contains(sb->dns_services, i, owner_family))
continue;
r = dns_service_split(i->ptr.name, &name, &type, &domain);
if (r < 0)
goto finish;
if (!name) {
type = mfree(type);
domain = mfree(domain);
r = dns_service_split(dns_resource_key_name(i->key), &name, &type, &domain);
if (r < 0)
goto finish;
}
if (!type)
continue;
r = dns_add_new_service(sb, i, owner_family);
if (r < 0)
goto finish;
log_debug("Add into the list %s, %s, %s, %s, %d",
strna(name),
strna(type),
strna(domain),
strna(af_to_ipv4_ipv6(owner_family)),
sb->ifindex);
r = sd_json_buildo(&entry,
SD_JSON_BUILD_PAIR("add_flag", SD_JSON_BUILD_BOOLEAN(true)),
SD_JSON_BUILD_PAIR("family", SD_JSON_BUILD_INTEGER(owner_family)),
SD_JSON_BUILD_PAIR("name", SD_JSON_BUILD_STRING(name?: "")),
SD_JSON_BUILD_PAIR("type", SD_JSON_BUILD_STRING(type?: "")),
SD_JSON_BUILD_PAIR("domain", SD_JSON_BUILD_STRING(domain?: "")),
SD_JSON_BUILD_PAIR("interface", SD_JSON_BUILD_INTEGER(sb->ifindex)));
if (r < 0)
goto finish;
r = sd_json_variant_append_array(&array, entry);
if (r < 0)
goto finish;
}
/* Check for services removed */
LIST_FOREACH(dns_services, service, sb->dns_services) {
_cleanup_free_ char *name = NULL, *type = NULL, *domain = NULL;
_cleanup_(sd_json_variant_unrefp) sd_json_variant *entry = NULL;
if (service->family != owner_family)
continue;
if (dns_answer_contains(answer, service->rr))
continue;
r = dns_service_split(service->rr->ptr.name, &name, &type, &domain);
if (r < 0)
goto finish;
if (!name) {
type = mfree(type);
domain = mfree(domain);
r = dns_service_split(dns_resource_key_name(service->rr->key), &name, &type, &domain);
if (r < 0)
goto finish;
}
dns_remove_service(sb, service);
log_debug("Remove from the list %s, %s, %s, %s, %d",
strna(name),
strna(type),
strna(domain),
strna(af_to_ipv4_ipv6(owner_family)),
sb->ifindex);
r = sd_json_buildo(&entry,
SD_JSON_BUILD_PAIR("add_flag", SD_JSON_BUILD_BOOLEAN(false)),
SD_JSON_BUILD_PAIR("family", SD_JSON_BUILD_INTEGER(owner_family)),
SD_JSON_BUILD_PAIR("name", SD_JSON_BUILD_STRING(name?: "")),
SD_JSON_BUILD_PAIR("type", SD_JSON_BUILD_STRING(type?: "")),
SD_JSON_BUILD_PAIR("domain", SD_JSON_BUILD_STRING(domain?: "")),
SD_JSON_BUILD_PAIR("interface", SD_JSON_BUILD_INTEGER(sb->ifindex)));
if (r < 0)
goto finish;
r = sd_json_variant_append_array(&array, entry);
if (r < 0)
goto finish;
}
if (!sd_json_variant_is_blank_array(array)) {
_cleanup_(sd_json_variant_unrefp) sd_json_variant *vm = NULL;
r = sd_json_buildo(&vm,
SD_JSON_BUILD_PAIR("browser_service_data", SD_JSON_BUILD_VARIANT(array)));
if (r < 0)
goto finish;
r = sd_varlink_notify(sb->link, vm);
if (r < 0)
goto finish;
}
return 0;
finish:
log_error_errno(r, "Failed to process received services: %m");
return sd_varlink_error_errno(sb->link, r);
}
int mdns_browser_lookup_cache(DnsServiceBrowser *sb, int owner_family) {
_cleanup_(dns_answer_unrefp) DnsAnswer *lookup_ret_answer = NULL;
DnsScope *scope;
int r;
assert(sb);
assert(sb->m);
scope = manager_find_scope_from_protocol(sb->m, sb->ifindex, DNS_PROTOCOL_MDNS, owner_family);
if (!scope)
return 0;
dns_cache_prune(&scope->cache);
r = dns_cache_lookup(
&scope->cache,
sb->key,
sb->flags,
NULL,
&lookup_ret_answer,
NULL,
NULL,
NULL);
if (r < 0)
return r;
return mdns_manage_services_answer(sb, lookup_ret_answer, owner_family);
}
int mdns_notify_browsers_goodbye(DnsScope *scope) {
DnsServiceBrowser *sb = NULL;
int r;
if (!scope)
return 0;
HASHMAP_FOREACH(sb, scope->manager->dns_service_browsers) {
r = mdns_browser_lookup_cache(sb, scope->family);
if (r < 0)
goto finish;
}
return 0;
finish:
return r;
}
int mdns_notify_browsers_unsolicited_updates(Manager *m, DnsAnswer *answer, int owner_family) {
DnsServiceBrowser *sb = NULL;
int r;
assert(m);
if (!answer)
return 0;
if (!m->dns_service_browsers)
return 0;
HASHMAP_FOREACH(sb, m->dns_service_browsers) {
r = dns_answer_match_key(answer, sb->key, NULL);
if (r < 0)
goto finish;
else if (r == 0)
continue;
r = mdns_browser_lookup_cache(sb, owner_family);
if (r < 0)
goto finish;
}
return 0;
finish:
return log_error_errno(r, "Failed to notify mDNS service subscribers, %m");
}
static void mdns_browse_service_query_complete(DnsQuery *q) {
_cleanup_(dns_service_browser_unrefp) DnsServiceBrowser *sb = NULL;
_cleanup_(dns_query_freep) DnsQuery *query = q;
int r;
assert(query);
assert(query->manager);
if (query->state != DNS_TRANSACTION_SUCCESS)
return;
sb = dns_service_browser_ref(hashmap_get(query->manager->dns_service_browsers, query->varlink_request));
if (!sb)
return;
r = dns_answer_match_key(query->answer, sb->key, NULL);
if (r < 0)
goto finish;
else if (r == 0)
return;
r = mdns_browser_lookup_cache(sb, query->answer_family);
if (r < 0)
goto finish;
/* When the query is answered from cache, we only get answers for one answer_family
* i.e. either ipv4 or ipv6.
* We need to perform another cache lookup for the other answer_family */
if (query->answer_query_flags == SD_RESOLVED_FROM_CACHE) {
r = mdns_browser_lookup_cache(sb, query->answer_family == AF_INET? AF_INET6 : AF_INET);
if (r < 0)
goto finish;
}
return;
finish:
log_error_errno(r, "mDNS browse query complete failed, %m");
}
static int mdns_next_query_schedule(sd_event_source *s, uint64_t usec, void *userdata) {
_cleanup_(dns_service_browser_unrefp) DnsServiceBrowser *sb = NULL;
_cleanup_(dns_query_freep) DnsQuery *q = NULL;
int r;
assert(userdata);
sb = dns_service_browser_ref(userdata);
r = dns_query_new(sb->m, &q, sb->question_utf8, sb->question_idna, NULL, sb->ifindex, sb->flags);
if (r < 0)
goto finish;
q->complete = mdns_browse_service_query_complete;
q->varlink_request = sd_varlink_ref(sb->link);
sd_varlink_set_userdata(sb->link, q);
r = dns_query_go(q);
if (r < 0)
goto finish;
/* RFC6762 5.2
* The intervals between successive queries MUST increase by at least a factor of two.
* When the interval between queries reaches or exceeds 60 minutes,perform
* subsequent queries at a steady-state rate of one query per hour */
if (sb->delay == 0) {
sb->delay++;
/* First query is sent wihtout SD_RESOLVED_NO_CACHE to fetch answers already in cache.
* Set SD_RESOLVED_NO_CACHE to make all subsequent queries go to the network. */
sb->flags |= SD_RESOLVED_NO_CACHE;
}
else
sb->delay = sb->delay < 2048 ? sb->delay * 2 : 3600;
r = event_reset_time_relative(
sb->m->event, &sb->schedule_event,
CLOCK_BOOTTIME, (sb->delay * USEC_PER_SEC),
0, mdns_next_query_schedule,
sb, 0, "mdns-next-query-schedule", true);
if (r < 0)
goto finish;
TAKE_PTR(q);
return 0;
finish:
return log_error_errno(r, "Failed to schedule mDNS query, %m");
}
void dns_service_browser_reset(Manager *m) {
int r;
if (!(m && m->dns_service_browsers))
return;
DnsServiceBrowser *sb;
HASHMAP_FOREACH(sb, m->dns_service_browsers) {
sb->delay = 0;
r = event_reset_time_relative(
sb->m->event, &sb->schedule_event,
CLOCK_BOOTTIME, (sb->delay * USEC_PER_SEC),
0, mdns_next_query_schedule,
sb, 0, "mdns-next-query-schedule", true);
if (r < 0)
log_error_errno(r, "Failed to reset mdns service subscriber, %m");
}
return;
}
int dns_subscribe_browse_service(
Manager *m,
sd_varlink *link,
const char *domain,
const char *name,
const char *type,
int ifindex,
uint64_t flags) {
_cleanup_(dns_service_browser_unrefp) DnsServiceBrowser *sb = NULL;
_cleanup_(dns_question_unrefp) DnsQuestion *question_idna = NULL, *question_utf8 = NULL;
int r;
assert(m);
assert(link);
if (ifindex <= 0)
return sd_varlink_error_invalid_parameter(link, JSON_VARIANT_STRING_CONST("ifindex"));
if (isempty(name))
name = NULL;
else if (!dns_service_name_is_valid(name))
return sd_varlink_error_invalid_parameter(link, JSON_VARIANT_STRING_CONST("name"));
if (isempty(type))
type = NULL;
else if (!dnssd_srv_type_is_valid(type))
return sd_varlink_error_invalid_parameter(link, JSON_VARIANT_STRING_CONST("type"));
r = dns_name_is_valid(domain);
if (r < 0)
return r;
if (r == 0)
return sd_varlink_error_invalid_parameter(link, JSON_VARIANT_STRING_CONST("domain"));
r = dns_question_new_service_type(&question_utf8, name, type, domain, false, DNS_TYPE_PTR);
if (r < 0)
return r;
r = dns_question_new_service_type(&question_idna, name, type, domain, true, DNS_TYPE_PTR);
if (r < 0)
return r;
sb = new(DnsServiceBrowser, 1);
if (!sb)
return log_oom();
*sb = (DnsServiceBrowser) {
.n_ref = 1,
.m = m,
.link = sd_varlink_ref(link),
.question_utf8 = dns_question_ref(question_utf8),
.question_idna = dns_question_ref(question_idna),
.key = dns_question_first_key(question_utf8),
.ifindex = ifindex,
.flags = flags,
};
/* Only mDNS continuous querying is currently supported. See RFC 6762 */
switch (flags & SD_RESOLVED_PROTOCOLS_ALL) {
case SD_RESOLVED_MDNS:
r = sd_event_add_time(m->event,
&sb->schedule_event,
CLOCK_BOOTTIME,
usec_add(now(CLOCK_BOOTTIME), (sb->delay * USEC_PER_SEC)),
0,
mdns_next_query_schedule,
sb);
if (r < 0)
return r;
break;
default:
return -EINVAL;
}
r = hashmap_ensure_put(&m->dns_service_browsers, NULL, link, sb);
if (r < 0)
return r;
TAKE_PTR(sb);
return 0;
}
DnsServiceBrowser *dns_service_browser_free(DnsServiceBrowser *sb) {
DnsQuery *q;
if (!sb)
return NULL;
LIST_FOREACH(dns_services, service, sb->dns_services)
dns_remove_service(sb, service);
sd_event_source_disable_unref(sb->schedule_event);
q = sd_varlink_get_userdata(sb->link);
if (q && DNS_TRANSACTION_IS_LIVE(q->state))
dns_query_complete(q, DNS_TRANSACTION_ABORTED);
dns_question_unref(sb->question_idna);
dns_question_unref(sb->question_utf8);
sd_varlink_unref(sb->link);
return mfree(sb);
}
DEFINE_TRIVIAL_REF_UNREF_FUNC(DnsServiceBrowser, dns_service_browser, dns_service_browser_free);

View File

@ -0,0 +1,79 @@
/* SPDX-License-Identifier: LGPL-2.1-or-later */
#pragma once
typedef struct DnsServiceBrowser DnsServiceBrowser;
#include "resolved-dns-query.h"
#include "resolved-manager.h"
#include "sd-varlink.h"
typedef struct DnsService DnsService;
typedef enum DnsRecordTTLState DnsRecordTTLState;
enum DnsRecordTTLState {
MDNS_TTL_80_PERCENT,
MDNS_TTL_85_PERCENT,
MDNS_TTL_90_PERCENT,
MDNS_TTL_95_PERCENT,
MDNS_TTL_100_PERCENT
};
struct DnsService {
unsigned n_ref;
DnsServiceBrowser *sb;
sd_event_source *schedule_event;
DnsResourceRecord *rr;
int family;
usec_t until;
DnsRecordTTLState rr_ttl_state;
DnsQuery *query;
LIST_FIELDS(DnsService, dns_services);
};
struct DnsServiceBrowser {
unsigned n_ref;
Manager *m;
sd_varlink *link;
DnsQuestion *question_idna;
DnsQuestion *question_utf8;
uint64_t flags;
sd_event_source *schedule_event;
usec_t delay;
DnsResourceKey *key;
int ifindex;
uint64_t token;
LIST_HEAD(DnsService, dns_services);
};
DnsServiceBrowser *dns_service_browser_free(DnsServiceBrowser *sb);
void dns_remove_service(DnsServiceBrowser *sb, DnsService *service);
DnsService *dns_service_free(DnsService *service);
DnsServiceBrowser* dns_service_browser_ref(DnsServiceBrowser *sb);
DnsServiceBrowser* dns_service_browser_unref(DnsServiceBrowser *sb);
DnsService* dns_service_ref(DnsService *service);
DnsService* dns_service_unref(DnsService *service);
void dns_browse_services_purge(Manager *m, int family);
void dns_service_browser_reset(Manager *m);
DEFINE_TRIVIAL_CLEANUP_FUNC(DnsServiceBrowser*, dns_service_browser_unref);
DEFINE_TRIVIAL_CLEANUP_FUNC(DnsService*, dns_service_unref);
bool dns_service_contains(DnsService *services, DnsResourceRecord *rr, int owner_family);
int mdns_manage_services_answer(DnsServiceBrowser *sb, DnsAnswer *answer, int owner_family);
int dns_add_new_service(DnsServiceBrowser *sb, DnsResourceRecord *rr, int owner_family);
int mdns_service_update(DnsService *service, DnsResourceRecord *rr, usec_t t);
int mdns_browser_lookup_cache(DnsServiceBrowser *sb, int owner_family);
int dns_subscribe_browse_service(Manager *m,
sd_varlink *link,
const char *domain,
const char * name,
const char * type,
int ifindex,
uint64_t flags);
int mdns_notify_browsers_unsolicited_updates(Manager *m, DnsAnswer *answer, int owner_family);
int mdns_notify_browsers_goodbye(DnsScope *scope);

View File

@ -1011,6 +1011,7 @@ static int answer_add_clamp_ttl(
}
}
rr->until = until;
r = dns_answer_add_extend(answer, rr, ifindex, answer_flags, rrsig);
if (r < 0)
return r;

View File

@ -416,6 +416,66 @@ int dns_question_new_reverse(DnsQuestion **ret, int family, const union in_addr_
return 0;
}
int dns_question_new_service_type(
DnsQuestion **ret,
const char *service,
const char *type,
const char *domain,
bool convert_idna,
uint16_t record_type) {
_cleanup_(dns_resource_key_unrefp) DnsResourceKey *key = NULL;
_cleanup_(dns_question_unrefp) DnsQuestion *q = NULL;
_cleanup_free_ char *buf = NULL, *joined = NULL;
const char *name;
int r;
assert(ret);
if (record_type == DNS_TYPE_SRV)
return -EINVAL;
if (!domain)
return -EINVAL;
if (type) {
if (convert_idna) {
r = dns_name_apply_idna(domain, &buf);
if (r < 0)
return r;
if (r > 0)
domain = buf;
}
r = dns_service_join(service, type, domain, &joined);
if (r < 0)
return r;
name = joined;
} else {
if (service)
return -EINVAL;
name = domain;
}
q = dns_question_new(1);
if (!q)
return -ENOMEM;
key = dns_resource_key_new(DNS_CLASS_IN, record_type, name);
if (!key)
return -ENOMEM;
r = dns_question_add(q, key, 0);
if (r < 0)
return r;
*ret = TAKE_PTR(q);
return 0;
}
int dns_question_new_service(
DnsQuestion **ret,
const char *service,

View File

@ -31,6 +31,7 @@ DnsQuestion *dns_question_unref(DnsQuestion *q);
int dns_question_new_address(DnsQuestion **ret, int family, const char *name, bool convert_idna);
int dns_question_new_reverse(DnsQuestion **ret, int family, const union in_addr_union *a);
int dns_question_new_service(DnsQuestion **ret, const char *service, const char *type, const char *domain, bool with_txt, bool convert_idna);
int dns_question_new_service_type(DnsQuestion **ret, const char *service, const char *type, const char *domain, bool convert_idna, uint16_t record_type);
int dns_question_add_raw(DnsQuestion *q, DnsResourceKey *key, DnsQuestionFlags flags);
int dns_question_add(DnsQuestion *q, DnsResourceKey *key, DnsQuestionFlags flags);

View File

@ -398,6 +398,7 @@ DnsResourceRecord* dns_resource_record_new(DnsResourceKey *key) {
.n_ref = 1,
.key = dns_resource_key_ref(key),
.expiry = USEC_INFINITY,
.until = USEC_INFINITY,
.n_skip_labels_signer = UINT8_MAX,
.n_skip_labels_source = UINT8_MAX,
};
@ -1704,6 +1705,7 @@ DnsResourceRecord *dns_resource_record_copy(DnsResourceRecord *rr) {
copy->ttl = rr->ttl;
copy->expiry = rr->expiry;
copy->until = rr->until;
copy->n_skip_labels_signer = rr->n_skip_labels_signer;
copy->n_skip_labels_source = rr->n_skip_labels_source;
copy->unparsable = rr->unparsable;

View File

@ -107,6 +107,8 @@ struct DnsResourceRecord {
unsigned n_ref;
uint32_t ttl;
usec_t expiry; /* RRSIG signature expiry */
usec_t until; /* Used to pass until of a record when doing a dns_cache_lookup().
* Needed to schedule cache maintenance queries when browsing for services. */
DnsResourceKey *key;

View File

@ -10,6 +10,7 @@
#include "hostname-util.h"
#include "missing_network.h"
#include "random-util.h"
#include "resolved-dns-browse-services.h"
#include "resolved-dnssd.h"
#include "resolved-dns-scope.h"
#include "resolved-dns-synthesize.h"
@ -121,6 +122,9 @@ DnsScope* dns_scope_free(DnsScope *s) {
dns_cache_flush(&s->cache);
dns_zone_flush(&s->zone);
/* Clear records of mDNS service browse subscriber, since cache bas been flushed */
dns_browse_services_purge(s->manager, s->family);
LIST_REMOVE(scopes, s->manager->dns_scopes, s);
return mfree(s);
}

View File

@ -14,6 +14,7 @@
#include "mkdir.h"
#include "netif-util.h"
#include "parse-util.h"
#include "resolved-dns-browse-services.h"
#include "resolved-link.h"
#include "resolved-llmnr.h"
#include "resolved-mdns.h"
@ -166,6 +167,7 @@ void link_allocate_scopes(Link *l) {
r = dns_scope_new(l->manager, &l->mdns_ipv4_scope, l, DNS_PROTOCOL_MDNS, AF_INET);
if (r < 0)
log_link_warning_errno(l, r, "Failed to allocate mDNS IPv4 scope, ignoring: %m");
dns_service_browser_reset(l->manager);
}
} else
l->mdns_ipv4_scope = dns_scope_free(l->mdns_ipv4_scope);
@ -176,6 +178,7 @@ void link_allocate_scopes(Link *l) {
r = dns_scope_new(l->manager, &l->mdns_ipv6_scope, l, DNS_PROTOCOL_MDNS, AF_INET6);
if (r < 0)
log_link_warning_errno(l, r, "Failed to allocate mDNS IPv6 scope, ignoring: %m");
dns_service_browser_reset(l->manager);
}
} else
l->mdns_ipv6_scope = dns_scope_free(l->mdns_ipv6_scope);

View File

@ -769,6 +769,7 @@ int manager_start(Manager *m) {
Manager *manager_free(Manager *m) {
Link *l;
DnssdService *s;
DnsServiceBrowser *sb;
if (!m)
return NULL;
@ -840,6 +841,10 @@ Manager *manager_free(Manager *m) {
dns_trust_anchor_flush(&m->trust_anchor);
manager_etc_hosts_flush(m);
while ((sb = hashmap_first(m->dns_service_browsers)))
dns_service_browser_free(sb);
hashmap_free(m->dns_service_browsers);
return mfree(m);
}
@ -1474,29 +1479,28 @@ bool manager_packet_from_our_transaction(Manager *m, DnsPacket *p) {
return t->sent && dns_packet_equal(t->sent, p);
}
DnsScope* manager_find_scope(Manager *m, DnsPacket *p) {
DnsScope* manager_find_scope_from_protocol(Manager *m, int ifindex, DnsProtocol protocol, int family) {
Link *l;
assert(m);
assert(p);
l = hashmap_get(m->links, INT_TO_PTR(p->ifindex));
l = hashmap_get(m->links, INT_TO_PTR(ifindex));
if (!l)
return NULL;
switch (p->protocol) {
switch (protocol) {
case DNS_PROTOCOL_LLMNR:
if (p->family == AF_INET)
if (family == AF_INET)
return l->llmnr_ipv4_scope;
else if (p->family == AF_INET6)
else if (family == AF_INET6)
return l->llmnr_ipv6_scope;
break;
case DNS_PROTOCOL_MDNS:
if (p->family == AF_INET)
if (family == AF_INET)
return l->mdns_ipv4_scope;
else if (p->family == AF_INET6)
else if (family == AF_INET6)
return l->mdns_ipv6_scope;
break;
@ -1706,6 +1710,9 @@ void manager_flush_caches(Manager *m, int log_level) {
LIST_FOREACH(scopes, scope, m->dns_scopes)
dns_cache_flush(&scope->cache);
dns_browse_services_purge(m, AF_UNSPEC); /* Clear records of DNS service browse subscriber, since caches are flushed */
dns_service_browser_reset(m);
log_full(log_level, "Flushed all caches.");
}

View File

@ -16,6 +16,7 @@
typedef struct Manager Manager;
#include "resolved-dns-browse-services.h"
#include "resolved-dns-query.h"
#include "resolved-dns-search-domain.h"
#include "resolved-dns-stream.h"
@ -161,6 +162,9 @@ struct Manager {
size_t n_socket_graveyard;
struct sigrtmin18_info sigrtmin18_info;
/* Map varlink links to DnsServiceBrowser instances. */
Hashmap *dns_service_browsers;
};
/* Manager */
@ -188,7 +192,13 @@ int manager_next_hostname(Manager *m);
bool manager_packet_from_local_address(Manager *m, DnsPacket *p);
bool manager_packet_from_our_transaction(Manager *m, DnsPacket *p);
DnsScope* manager_find_scope(Manager *m, DnsPacket *p);
DnsScope* manager_find_scope_from_protocol(Manager *m, int ifindex, DnsProtocol protocol, int family);
static inline DnsScope* manager_find_scope(Manager *m, DnsPacket *p) {
assert(m);
assert(p);
return manager_find_scope_from_protocol(m, p->ifindex, p->protocol, p->family);
}
void manager_verify_all(Manager *m);

View File

@ -359,6 +359,10 @@ static int mdns_goodbye_callback(sd_event_source *s, uint64_t usec, void *userda
dns_cache_prune(&scope->cache);
r = mdns_notify_browsers_goodbye(scope);
if (r < 0)
log_error_errno(r, "mDNS: Failed to notify service subscribers of goodbyes, %m");
if (dns_cache_expiry_in_one_second(&scope->cache, usec)) {
r = sd_event_add_time_relative(
scope->manager->event,
@ -380,6 +384,7 @@ static int on_mdns_packet(sd_event_source *s, int fd, uint32_t revents, void *us
Manager *m = userdata;
DnsScope *scope;
int r;
bool unsolicited_packet = true;
r = manager_recv(m, fd, DNS_PROTOCOL_MDNS, &p);
if (r <= 0)
@ -455,30 +460,6 @@ static int on_mdns_packet(sd_event_source *s, int fd, uint32_t revents, void *us
}
}
for (bool match = true; match;) {
match = false;
LIST_FOREACH(transactions_by_scope, t, scope->transactions) {
if (t->state != DNS_TRANSACTION_PENDING)
continue;
r = dns_answer_match_key(p->answer, dns_transaction_key(t), NULL);
if (r <= 0) {
if (r < 0)
log_debug_errno(r, "Failed to match resource key, ignoring: %m");
continue;
}
/* This packet matches the transaction, let's pass it on as reply */
dns_transaction_process_reply(t, p, false);
/* The dns_transaction_process_reply() -> dns_transaction_complete() ->
* dns_query_candidate_stop() may free multiple transactions. Hence, restart
* the loop. */
match = true;
break;
}
}
dns_cache_put(
&scope->cache,
scope->manager->enable_cache,
@ -494,6 +475,35 @@ static int on_mdns_packet(sd_event_source *s, int fd, uint32_t revents, void *us
&p->sender,
scope->manager->stale_retention_usec);
for (bool match = true; match;) {
match = false;
LIST_FOREACH(transactions_by_scope, t, scope->transactions) {
if (t->state != DNS_TRANSACTION_PENDING)
continue;
r = dns_answer_match_key(p->answer, dns_transaction_key(t), NULL);
if (r <= 0) {
if (r < 0)
log_debug_errno(r, "Failed to match resource key, ignoring: %m");
continue;
}
unsolicited_packet = false;
/* This packet matches the transaction, let's pass it on as reply */
dns_transaction_process_reply(t, p, false);
/* The dns_transaction_process_reply() -> dns_transaction_complete() ->
* dns_query_candidate_stop() may free multiple transactions. Hence, restart
* the loop. */
match = true;
break;
}
}
/* Check incoming packet key matches with active clients if yes update the same */
if (unsolicited_packet)
mdns_notify_browsers_unsolicited_updates(m, p->answer, p->family);
} else if (dns_packet_validate_query(p) > 0) {
log_debug("Got mDNS query packet for id %u", DNS_PACKET_ID(p));

View File

@ -4,6 +4,7 @@
#include "glyph-util.h"
#include "in-addr-util.h"
#include "json-util.h"
#include "resolved-dns-browse-services.h"
#include "resolved-dns-synthesize.h"
#include "resolved-varlink.h"
#include "socket-netlink.h"
@ -30,11 +31,26 @@ typedef struct LookupParametersResolveService {
uint64_t flags;
} LookupParametersResolveService;
typedef struct LookupParametersMdnsBrowse {
char *domainName;
char *name;
char *type;
int ifindex;
uint64_t flags;
} LookupParametersMdnsBrowse;
static void lookup_parameters_destroy(LookupParameters *p) {
assert(p);
free(p->name);
}
static void lookup_parameters_mdns_destroy(LookupParametersMdnsBrowse *p) {
assert(p);
free(p->domainName);
free(p->name);
free(p->type);
}
static int reply_query_state(DnsQuery *q) {
assert(q);
@ -108,10 +124,18 @@ static int reply_query_state(DnsQuery *q) {
static void vl_on_disconnect(sd_varlink_server *s, sd_varlink *link, void *userdata) {
DnsQuery *q;
Manager *m;
assert(s);
assert(link);
m = sd_varlink_server_get_userdata(s);
if (!m)
return;
DnsServiceBrowser *sb = hashmap_remove(m->dns_service_browsers, link);
dns_service_browser_free(sb);
q = sd_varlink_get_userdata(link);
if (!q)
return;
@ -1209,6 +1233,43 @@ static int verify_polkit(sd_varlink *link, sd_json_variant *parameters, const ch
&m->polkit_registry);
}
static int vl_method_start_browse(sd_varlink* link, sd_json_variant* parameters, sd_varlink_method_flags_t flags, void* userdata) {
static const sd_json_dispatch_field dispatch_table[] = {
{ "domainName", SD_JSON_VARIANT_STRING, sd_json_dispatch_string, offsetof(LookupParametersMdnsBrowse, domainName), SD_JSON_MANDATORY },
{ "name", SD_JSON_VARIANT_STRING, sd_json_dispatch_string, offsetof(LookupParametersMdnsBrowse, name), 0 },
{ "type", SD_JSON_VARIANT_STRING, sd_json_dispatch_string, offsetof(LookupParametersMdnsBrowse, type), 0 },
{ "ifindex", _SD_JSON_VARIANT_TYPE_INVALID, sd_json_dispatch_int, offsetof(LookupParametersMdnsBrowse, ifindex), SD_JSON_MANDATORY },
{ "flags", _SD_JSON_VARIANT_TYPE_INVALID, sd_json_dispatch_uint64, offsetof(LookupParametersMdnsBrowse, flags), SD_JSON_MANDATORY },
{}
};
_cleanup_(lookup_parameters_mdns_destroy) LookupParametersMdnsBrowse p = {};
Manager *m;
int r = 0;
assert(link);
/* if the client didn't set the more flag, it is using us incorrectly */
if (!FLAGS_SET(flags, SD_VARLINK_METHOD_MORE))
return sd_varlink_error(link, SD_VARLINK_ERROR_EXPECTED_MORE, NULL);
m = sd_varlink_server_get_userdata(sd_varlink_get_server(link));
assert(m);
r = sd_varlink_dispatch(link, parameters, dispatch_table, &p);
if (r < 0)
return log_error_errno(r, "vl_method_start_browse json_dispatch fail: %m");
if (!validate_and_mangle_flags(NULL, &p.flags, 0))
return sd_varlink_error_invalid_parameter(link, JSON_VARIANT_STRING_CONST("flags"));
r = dns_subscribe_browse_service(m, link, p.domainName, p.name, p.type, p.ifindex, p.flags);
if (r < 0)
return sd_varlink_error_errno(link, r);
return 1;
}
static int vl_method_subscribe_query_results(sd_varlink *link, sd_json_variant *parameters, sd_varlink_method_flags_t flags, void *userdata) {
Manager *m = ASSERT_PTR(sd_varlink_get_userdata(ASSERT_PTR(link)));
int r;
@ -1422,7 +1483,8 @@ static int varlink_main_server_init(Manager *m) {
"io.systemd.Resolve.ResolveHostname", vl_method_resolve_hostname,
"io.systemd.Resolve.ResolveAddress", vl_method_resolve_address,
"io.systemd.Resolve.ResolveService", vl_method_resolve_service,
"io.systemd.Resolve.ResolveRecord", vl_method_resolve_record);
"io.systemd.Resolve.ResolveRecord", vl_method_resolve_record,
"io.systemd.Resolve.StartBrowse", vl_method_start_browse);
if (r < 0)
return log_error_errno(r, "Failed to register varlink methods: %m");

View File

@ -102,6 +102,15 @@ static SD_VARLINK_DEFINE_STRUCT_TYPE(
SD_VARLINK_DEFINE_FIELD(ifindex, SD_VARLINK_INT, SD_VARLINK_NULLABLE),
SD_VARLINK_DEFINE_FIELD(name, SD_VARLINK_STRING, 0));
static SD_VARLINK_DEFINE_STRUCT_TYPE(
ServiceData,
SD_VARLINK_DEFINE_FIELD(add_flag, SD_VARLINK_BOOL, 0),
SD_VARLINK_DEFINE_FIELD(family, SD_VARLINK_INT, 0),
SD_VARLINK_DEFINE_FIELD(name, SD_VARLINK_STRING, SD_VARLINK_NULLABLE),
SD_VARLINK_DEFINE_FIELD(type, SD_VARLINK_STRING, 0),
SD_VARLINK_DEFINE_FIELD(domain, SD_VARLINK_STRING, 0),
SD_VARLINK_DEFINE_FIELD(interface, SD_VARLINK_INT, 0));
static SD_VARLINK_DEFINE_METHOD(
ResolveAddress,
SD_VARLINK_FIELD_COMMENT("The Linux interface index for the network interface to search on. Typically left unspecified, in order to search on all interfaces."),
@ -159,6 +168,15 @@ static SD_VARLINK_DEFINE_METHOD(
SD_VARLINK_DEFINE_OUTPUT_BY_TYPE(rrs, ResolvedRecord, SD_VARLINK_ARRAY),
SD_VARLINK_DEFINE_OUTPUT(flags, SD_VARLINK_INT, 0));
static SD_VARLINK_DEFINE_METHOD(
StartBrowse,
SD_VARLINK_DEFINE_INPUT(domainName, SD_VARLINK_STRING, 0),
SD_VARLINK_DEFINE_INPUT(name, SD_VARLINK_STRING, SD_VARLINK_NULLABLE),
SD_VARLINK_DEFINE_INPUT(type, SD_VARLINK_STRING, SD_VARLINK_NULLABLE),
SD_VARLINK_DEFINE_INPUT(ifindex, SD_VARLINK_INT, 0),
SD_VARLINK_DEFINE_INPUT(flags, SD_VARLINK_INT, 0),
SD_VARLINK_DEFINE_OUTPUT_BY_TYPE(browser_service_data, ServiceData, SD_VARLINK_ARRAY));
static SD_VARLINK_DEFINE_ERROR(NoNameServers);
static SD_VARLINK_DEFINE_ERROR(NoSuchResourceRecord);
static SD_VARLINK_DEFINE_ERROR(QueryTimedOut);
@ -198,6 +216,8 @@ SD_VARLINK_DEFINE_INTERFACE(
&vl_method_ResolveService,
SD_VARLINK_SYMBOL_COMMENT("Resolves a domain name to one or more DNS resource records."),
&vl_method_ResolveRecord,
SD_VARLINK_SYMBOL_COMMENT("Starts browsing for mDNS services of specified type."),
&vl_method_StartBrowse,
SD_VARLINK_SYMBOL_COMMENT("Encapsulates a resolved address."),
&vl_type_ResolvedAddress,
SD_VARLINK_SYMBOL_COMMENT("Encapsulates a resolved host name."),
@ -212,6 +232,7 @@ SD_VARLINK_DEFINE_INTERFACE(
&vl_type_ResourceRecord,
SD_VARLINK_SYMBOL_COMMENT("Encapsulates information about a resolved DNS resource record "),
&vl_type_ResolvedRecord,
&vl_type_ServiceData,
&vl_error_NoNameServers,
&vl_error_NoSuchResourceRecord,
&vl_error_QueryTimedOut,

View File

@ -522,6 +522,7 @@ static int reply_callback(
if (!arg_quiet)
sd_json_variant_dump(parameters, arg_json_format_flags, stdout, NULL);
fflush(stdout);
return r;
}

View File

@ -0,0 +1,6 @@
# SPDX-License-Identifier: LGPL-2.1-or-later
all setup run clean clean-again:
@TEST_BASE_DIR=../ ./test.sh --$@
.PHONY: all setup run clean clean-again

View File

@ -0,0 +1,8 @@
# SPDX-License-Identifier: LGPL-2.1-or-later
integration_tests += [
integration_test_template + {
'name' : fs.name(meson.current_source_dir()),
'vm' : true,
},
]

View File

@ -0,0 +1,12 @@
#!/usr/bin/env bash
# SPDX-License-Identifier: LGPL-2.1-or-later
set -e
TEST_DESCRIPTION="Test for systemd-resolved's mDNS functionality"
IMAGE_NAME="resolved-mdns"
TEST_NO_NSPAWN=1
# shellcheck source=test/test-functions
. "${TEST_BASE_DIR:?}/test-functions"
do_test "$@"

View File

@ -377,6 +377,7 @@ foreach dirname : [
'TEST-84-STORAGETM',
'TEST-85-NETWORK',
'TEST-86-MULTI-PROFILE-UKI',
'TEST-87-RESOLVED-MDNS',
]
subdir(dirname)
endforeach

View File

@ -80,6 +80,7 @@ show_journal = True # When true, show journal on stopping networkd.
active_units = []
protected_links = {
'br0',
'erspan0',
'gre0',
'gretap0',

View File

@ -0,0 +1,8 @@
# SPDX-License-Identifier: LGPL-2.1-or-later
[Unit]
Description=TEST-87-RESOLVED-MDNS
[Service]
ExecStartPre=rm -f /failed /testok
ExecStart=/usr/lib/systemd/tests/testdata/units/%N.sh
Type=oneshot

View File

@ -0,0 +1,255 @@
#!/usr/bin/env bash
# SPDX-License-Identifier: LGPL-2.1-or-later
set -eux
set -o pipefail
# shellcheck source=test/units/test-control.sh
. "$(dirname "$0")"/test-control.sh
SERVICE_TYPE_COUNT=10
SERVICE_COUNT=20
CONTAINER_ZONE="test-$RANDOM"
CONTAINER_1="test-mdns-1"
CONTAINER_2="test-mdns-2"
# Prepare containers
create_container() {
local container="${1:?}"
local stype sid svc
# Prepare container's /etc
#
# Since we also need the various test suite related dropins from the host's /etc,
# we'll overlay our customizations on top of that
mkdir -p "/var/lib/machines/$container/etc/systemd/dnssd"
# Create 20 test services for each service type (_testServiceX._udp) and number them sequentially,
# i.e. create services 0-19 for _testService0._udp, services 20-39 for _testService1._udp, and so on
for stype in $(seq 0 $((SERVICE_TYPE_COUNT - 1))); do
for sid in $(seq 0 $((SERVICE_COUNT - 1))); do
svc=$((stype * SERVICE_COUNT + sid))
cat >"/var/lib/machines/$container/etc/systemd/dnssd/test-service-$container-$svc.dnssd" <<EOF
[Service]
Name=Test Service $svc on %H
Type=_testService$stype._udp
Port=24002
TxtText=DC=Monitor PN=867313 SN=XZ051Z0051
EOF
done
done
# To make things fast, spawn the container with a transient version of what's currently the host's
# rootfs, sans a couple of tweaks to make the container unique enough
mkdir -p "/run/systemd/system/systemd-nspawn@$container.service.d"
cat >"/run/systemd/system/systemd-nspawn@$container.service.d/override.conf" <<EOF
[Service]
ExecStart=
ExecStart=systemd-nspawn --quiet --link-journal=try-guest --keep-unit --machine=%i --boot \
--volatile=yes --directory=/ \
--inaccessible=/etc/machine-id \
--inaccessible=/etc/hostname \
--resolv-conf=replace-stub \
--network-zone=$CONTAINER_ZONE \
--overlay=/etc:/var/lib/machines/$container/etc::/etc \
--hostname=$container
EOF
}
check_both() {
local service_id="${1:?}"
local result_file="${2:?}"
# We should get 20 services per container, 40 total
if [[ "$(wc -l <"$result_file")" -ge 40 ]]; then
# Check if the services we got are the correct ones
for i in $(seq 0 $((SERVICE_TYPE_COUNT - 1))); do
svc=$((service_id * SERVICE_COUNT + i))
if ! grep "Test Service $svc on $CONTAINER_1" "$result_file" || \
! grep "Test Service $svc on $CONTAINER_2" "$result_file"; then
return 1
fi
done
# We got all records and all of them are what we expect
return 0
fi
return 1
}
check_first() {
local service_id="${1:?}"
local result_file="${2:?}"
# We should get 20 services per container
if [[ "$(wc -l <"$result_file")" -ge 20 ]]; then
# Check if the services we got are the correct ones
for i in $(seq 0 $((SERVICE_TYPE_COUNT - 1))); do
svc=$((service_id * SERVICE_COUNT + i))
if ! grep "Test Service $svc on $CONTAINER_1" "$result_file"; then
return 1
fi
# This check assumes the second container is unreachable, so this shouldn't happen
if grep "Test Service $svc on $CONTAINER_2" "$result_file"; then
echo >&2 "Found a record from an unreachable container"
cat "$result_file"
exit 1
fi
done
# We got all records and all of them are what we expect
return 0
fi
return 1
}
run_and_check_services() {
local service_id="${1:?}"
local check_func="${2:?}"
local unit_name="varlinkctl-$service_id-$SRANDOM.service"
local i out_file parameters service_type svc tmp_file
out_file="$(mktemp)"
error_file="$(mktemp)"
tmp_file="$(mktemp)"
service_type="_testService$service_id._udp"
parameters="{ \"domainName\": \"$service_type.local\", \"name\": \"\", \"type\": \"\", \"ifindex\": ${BRIDGE_INDEX:?}, \"flags\": 16785432 }"
systemd-run --unit="$unit_name" --service-type=exec -p StandardOutput="file:$out_file" -p StandardError="file:$error_file" \
varlinkctl call --more /run/systemd/resolve/io.systemd.Resolve io.systemd.Resolve.StartBrowse "$parameters"
# shellcheck disable=SC2064
# Note: unregister the trap once it's fired, otherwise it'll get propagated to functions that call this
# one, *sigh*
trap "trap - RETURN; systemctl stop $unit_name" RETURN
for _ in {0..14}; do
# The response format, for reference (it's JSON-SEQ):
#
# {
# "browser_service_data": [
# {
# "add_flag": true,
# "family": 10,
# "name": "Test Service 13 on test-mdns-1",
# "type": "_testService0._udp",
# "domain": "local",
# "interface": 3
# },
# ...
# ]
# }
if [[ -s "$out_file" ]]; then
# Extract the service name from each valid record...
# jq --slurp --raw-output \
# ".[].browser_service_data[] | select(.add_flag == true and .type == \"$service_type\" and .family == 10).name" "$out_file" | sort | tee "$tmp_file"
grep -o '"name":"[^"]*"' "$out_file" | sed 's/"name":"//;s/"//g' | sort | tee "$tmp_file"
# ...and compare them with what we expect
if "$check_func" "$service_id" "$tmp_file"; then
return 0
fi
fi
sleep 2
done
cat "$out_file"
cat "$error_file"
return 1
}
testcase_all_sequential() {
: "Test each service type (sequentially)"
resolvectl flush-caches
for id in $(seq 0 $((SERVICE_TYPE_COUNT - 1))); do
run_and_check_services "$id" check_both
done
echo testcase_end
}
testcase_all_parallel() {
: "Test each service type (in parallel)"
resolvectl flush-caches
for id in $(seq 0 $((SERVICE_TYPE_COUNT - 1))); do
run_and_check_services "$id" check_both &
done
wait
}
testcase_single_service_multiple_times() {
: "Test one service type multiple times"
resolvectl flush-caches
for _ in {0..4}; do
run_and_check_services 4 check_both
done
}
testcase_second_unreachable() {
: "Test each service type while the second container is unreachable"
resolvectl flush-caches
systemd-run -M "$CONTAINER_2" --wait --pipe -- networkctl down host0
for id in $(seq 0 $((SERVICE_TYPE_COUNT - 1))); do
run_and_check_services "$id" check_first
done
: "Test each service type after bringing the second container back up again"
systemd-run -M "$CONTAINER_2" --wait --pipe -- networkctl up host0
systemd-run -M "$CONTAINER_2" --wait --pipe -- \
/usr/lib/systemd/systemd-networkd-wait-online --ipv4 --ipv6 --interface=host0 --operational-state=degraded --timeout=30
for id in $(seq 0 $((SERVICE_TYPE_COUNT - 1))); do
run_and_check_services "$id" check_both
done
}
: "Setup host & containers"
# Note: create the drop-in intentionally under /run/ and copy it manually into the containers
mkdir -p /run/systemd/resolved.conf.d/
cat >/run/systemd/resolved.conf.d/99-mdns-llmnr.conf <<EOF
[Resolve]
MulticastDNS=yes
LLMNR=yes
EOF
systemctl unmask systemd-resolved.service systemd-networkd.{service,socket} systemd-machined.service
systemctl enable --now systemd-resolved.service systemd-networkd.{socket,service} systemd-machined.service
ln -svrf /run/systemd/resolved.conf.d/99-mdns-llmnr.conf /etc/resolv.conf
systemctl reload systemd-resolved.service systemd-networkd.service
for container in "$CONTAINER_1" "$CONTAINER_2"; do
create_container "$container"
mkdir -p "/var/lib/machines/$container/etc/systemd/resolved.conf.d/"
cp /run/systemd/resolved.conf.d/99-mdns-llmnr.conf "/var/lib/machines/$container/etc/systemd/resolved.conf.d/"
touch "/var/lib/machines/$container/etc/hostname"
systemctl daemon-reload
machinectl start "$container"
# Wait for the system bus to start...
timeout 30s bash -xec "while ! systemd-run -M '$container' --wait --pipe true; do sleep 1; done"
# ...and from there wait for the machine bootup to finish. We don't really care if the container
# boots up in a degraded state, hence the `:`
timeout 30s systemd-run -M "$container" --wait --pipe -- systemctl --wait is-system-running || :
# Wait until the veth interface is configured and turn on mDNS and LLMNR
systemd-run -M "$container" --wait --pipe -- \
/usr/lib/systemd/systemd-networkd-wait-online --ipv4 --ipv6 --interface=host0 --operational-state=degraded --timeout=30
systemd-run -M "$container" --wait --pipe -- resolvectl mdns host0 yes
systemd-run -M "$container" --wait --pipe -- resolvectl llmnr host0 yes
systemd-run -M "$container" --wait --pipe -- networkctl status --no-pager
systemd-run -M "$container" --wait --pipe -- resolvectl status --no-pager
[[ "$(systemd-run -M "$container" --wait --pipe -- resolvectl mdns host0)" =~ :\ yes$ ]]
[[ "$(systemd-run -M "$container" --wait --pipe -- resolvectl llmnr host0)" =~ :\ yes$ ]]
done
BRIDGE_INDEX="$(<"/sys/class/net/vz-$CONTAINER_ZONE/ifindex")"
machinectl list
resolvectl mdns "vz-$CONTAINER_ZONE" on
resolvectl llmnr "vz-$CONTAINER_ZONE" on
networkctl status
resolvectl status
# Run the actual test cases (functions prefixed by testcase_)
run_testcases
touch /testok