CHANGES WITH 260:

Feature Removals and Incompatible Changes:

    * Support for System V service scripts has been removed. Please make
      sure to update your software *now* to include a native systemd unit
      file instead of a legacy System V script.

      The following components have been removed:

      • systemd-rc-local-generator and rc-local.service,
      • systemd-sysv-generator,
      • systemd-sysv-install (hook for systemctl enable/disable/is-enabled).

      The corresponding meson options '-Drc-local=', '-Dsysvinit-path=',
      and '-Dsysvrcnd-path=' are deprecated, and will be dropped in a future
      release.


    * Meson options '-Dintegration-tests=' and '-Dcryptolib=' (deprecated
      in v258) have been removed.


    * Support for libidn has been removed. IDN functionality now requires
      libidn2. The corresponding meson option '-Dlibidn=' is deprecated
      too and will be dropped in a future release.


    * Required versions of various dependencies have been raised:

      • cryptsetup 2.0.1/2.3.0 → 2.4.0,
      • elfutils 158 → 177,
      • libblkid 2.24 → 2.37,
      • libseccomp 2.3.1 → 2.4.0,
      • glibc 2.31 → 2.34,
      • libxcrypt or libcrypt from glibc → libxcrypt 4.4.0 only,
      • OpenSSL 1.1.0 → 3.0.0,
      • Python 3.7.0 → 3.9.0.


    * The Linux kernel version requirements have been updated too:
      baseline 5.4 → 5.10, recommended baseline 5.7 → 5.14, 6.6 for full
      functionality. Code for compatibility with versions older than the
      baseline has been removed.


    * The parsing of RootImageOptions= and the mount image parameters of
      ExtensionImages= and MountImages= has been changed so that the last
      definition for a given partition wins and is applied, rather than the
      first, to keep these options coherent with other unit settings.


    * Support for non-system users and groups in udev rules and
      systemd-networkd configuration has been restored, but is deprecated
      and discouraged. systemd-udevd will emits warnings if a non-system
      user/group is specified in OWNER=/GROUP=. Similarly, systemd-networkd
      will warn about User=/Group= settings with a non-system user/group
      specified in .netdev files for Tun/Tap interfaces. This support will
      be removed in a future release.

      Device nodes should not be owned by a non-system user/group. It is
      recommended to check udev rules files with 'udevadm verify' and/or
      'udevadm test' commands.


    * systemd-repart will now make use of mkfs.xfs's support for
      populating XFS filesystems from a directory. This support was
      added in xfsprogs 6.17.0 released 20 October 2025. As there is no
      proper way to detect whether mkfs.xfs supports populating from a
      directory or not, we make use of it unconditionally and have dropped
      support for the old way using protofiles.


    * The org.systemd.login1.Manager D-Bus interface has a minor API break.
      The CanPowerOff(), CanReboot(), CanSuspend(), etc. family of methods
      have introduced new return values which may break downstream
      consumers such as desktop environments. The new return values more
      precisely communicate the status of inhibitors: 'inhibited',
      'inhibitor-blocked', and 'challenge-inhibitor-blocked'. This allows
      desktops to differentiate between system administrator policy and
      temporary restrictions imposed by inhibitors.


    * In systemd-260-rc1, the sd_varlink_field_type_t enum was extended in
      a way that changed the numerical values of existing fields. This was
      reverted for -rc2. Programs using sd-varlink and compiled with the
      headers from -rc1 must be recompiled.

New system interfaces and components:

    * The os-release(5) gained a new field FANCY_NAME= that is similar to
      PRETTY_NAME= but may contain ANSI sequences, and non-ASCII Unicode
      glyphs. The new field is also defined to NOT contain any version
      specification, providing better separation between the OS name and
      version.

      The systemd manager, systemd-hostnamed, and hostnamectl will now
      show FANCY_NAME= in preference to PRETTY_NAME=.


    * The "Portability and Stability" policy has been simplified and
      updated to strengthen the promises of avoiding user-visible
      regressions in public interfaces. See
      https://systemd.io/PORTABILITY_AND_STABILITY/ for details.


    * Services providing a public Varlink interface can be symlinked under
      /run/varlink/registry/, allowing well-known services to be
      enumerated. 'varlinkctl list-registry' can be used to list available
      services. This is particularly useful in context of the Varlink HTTP
      bridge (https://github.com/mvo5/varlink-http-bridge), which may
      expose all services whose sockets are linked in this directory.


    * A new "metrics" or "report" framework has been defined. Any system
      component can hook into the reporting framework by providing a
      Varlink endpoint under /run/systemd/report/.

      systemd-report is a new command line tool which collects the reports
      from all endpoints and combines them in JSON format.

      The details of the structure of the reports should be considered
      EXPERIMENTAL at this point. We reserve the right to make incompatible
      changes to the JSON structure and/or place additional requirements.

      Currently, two components provide metrics this way: systemd-networkd
      and the system service manager.


    * A new "mstack" feature has been introduced, to allowing defining an
      overlayfs and bind mount arrangement by structuring the content of an
      ".mstack/" directory that follows this specification. MStacks are
      useful to invoke services and containers from a directory that fully
      self describes its intended way of use.


    * A new 'verity' TPM NvPCR has been added. Various components measure
      dm-verity images to it upon loading. This includes
      systemd-veritysetup (controllable via the new 'tpm2-measure-nvpcr='
      /etc/veritytab setting), and the DDI dissection logic.


    * A canonical set of hwid files for automated DeviceTree mapping in
      UKIs is now shipped under /usr/lib/systemd/boot/hwids/<efi-arch>/.
      Users and developers are welcome to contribute definitions for
      their specific devices, so that UKIs can automatically find and load
      the appropriate DTB on boot, without requiring per-devices and
      per-kernel images. ukify now uses these definition automatically if
      present on the system at build time. A first set of hwid files for
      arm64 Snapdragon devices has been imported.

Changes in the system and service manager:

    * A new unit setting RootMStack= has been introduced, to support the
      new "mstack" feature for services (see above).


    * The unit setting PrivateUsers= gained a new possible value "managed",
      which automatically assigns a dynamic and transient range of 65536
      UIDs/GIDs to the unit, acquired via systemd-nsresourced.


    * The implementation for PrivateUsers=full has been updated to map the
      full range of IDs. The workaround to allow nested systemd older than
      257 to correctly detect that it is under such a mapping has been
      dropped.


    * systemd now uses the CSI 18 terminal sequence to query terminal
      size. This allows the query to be made without changing the position
      of the cursor. Terminal emulators which do not yet support the
      sequence are encouraged to do so.


    * Service units gained a RefreshOnReload= setting that configures
      whether extensions and credentials are to be refreshed when the unit
      is reloaded.


    * A new unit setting BindNetworkInterface= has been introduced that
      automatically binds all sockets created by the unit to a specific
      network interface. This is generally useful, but in particular for
      VRF setups.


    * Two new unit settings ConditionPathIsSocket= and AssertPathIsSocket=
      can be used to skip or fail the unit if the given path is not a
      socket.


    * For units which specify PrivateTmp=yes and DefaultDependencies=no
      without an explicit requirement for /tmp/, a disconnected /tmp/ will
      be used, as if PrivateTmp=disconnected was specified. Also, if there
      is no explicit ordering for /var/, the private mount for /var/tmp/
      will not be created. Those changes avoid race conditions with
      creation of those private directories during early boot and may
      result in changes to unit ordering.


    * EnqueueMarkedJobs() D-Bus method now has a Varlink counterpart.


    * systemctl gained a new 'enqueue-marked' verb, which calls the
      EnqueueMarkedJobs() D-Bus method. The '--marked' parameter, which was
      previously used for the same purpose, is now deprecated.


    * SetProperties() D-Bus method now has a Varlink counterpart. For now,
      it only supports setting the Markers= property.


    * New 'needs-start' and 'needs-stop' settings are now supported for the
      Markers= property.


    * The CPUSchedulingPolicy= service setting now supports the new value
      "ext" for enabling the SCHED_EXT scheduler recently added to the
      Linux kernel.


    * A new MemoryTHP= service setting has been added that controls
      per-service Transparent Huge Pages (THP) support.

## Changes in systemd-udevd:

    * Permissions for /dev/ptp* are now set to 0664 (previously 0660),
      allowing unprivileged read-only access. This relies on the kernel fix
      "ptp: Add PHC file mode checks. Allow RO adjtime() without
      FMODE_WRITE." (commit b4e53b15c04e3852949003752f48f7a14ae39e86 in
      v6.15, backported to LTS releases in v6.12.68, v6.6.122, v6.1.162,
      v5.15.199, and v5.10.249), which adds missing PTP ioctl permission
      checks and keeps clock-modifying operations write-restricted. Systems
      running stable kernel branches should ensure they are updated to patch
      levels that include the fix.


    * Persistent network interface naming has bee extended to MCTP devices
      with the "mc" prefix.


    * The minimum backlight brightness value used when restoring backlight
      levels at boot has been lowered from 5% to 1%. This lower value
      should be sufficient to avoid blacked-out displays, but allows user
      environments to use a wider range of values (without lower values
      being reset during reboot). Note that environments may still set very
      low brightness values at runtime independently of the systemd clamp
      which only applies during boot.


    * A new udev property ID_INTEGRATION= is now exposed on devices that
      have ID_BUS= defined. This variable can be set to 'internal' when the
      device is integral part of the system or 'external' otherwise.
      Internal buses like PCI, I2C, SPI... imply 'internal' and external
      buses like bluetooth imply 'external'. For USB the 'removable'
      attribute of the port the device is connected to determines the
      result: 'fixed' implies 'internal' and 'removable' or 'unknown'
      implies 'external'.


    * ID_INPUT_JOYSTICK_INTEGRATION= property has been dropped in favour of
      ID_INTEGRATION= because it was never used and the new variable covers
      the idea that variable was intended for better.


    * A new udev builtin "tpm2_id" is now available which will extract
      vendor/model identification from connected TPM2 devices as they are
      probed. This is then used to import data from the udev database,
      possibly containing quirk and other information about specific TPMs.

Changes in systemd-networkd:

    * MultiPathRoute= option now supports interface-bound ECMP routes.


    * systemd-networkd gained integration with ModemManager via the "simple
      connect" protocol. A new [MobileNetwork] section has been added with
      APN=, AllowedAuthenticationMechanisms=, User=, Password=, IPFamily=,
      AllowRoaming=, PIN=, OperatorId=, RouteMetric=, and UseGateway=
      settings. This allows systemd-networkd to establish a cellular modem
      connection to a broadband network.


    * systemd-networkd gained a pair of varlink methods
      io.systemd.Network.Link.Up()/Down(). 'networkctl up/down' now
      utilizes those varlink interfaces in place of direct RTNL messages
      for better interaction with networkd.


    * .link files gained new ScatterGather=, ScatterGatherFragmentList=,
      TCPECNSegmentationOffload=, TCPMangleIdSegmentationOffload=,
      GenericReceiveOffloadList=, GenericReceiveOffloadUDPForwarding=
      options for configuring various details of Ethernet devices.


    * systemd-networkd's Varlink and JSON interfaces will now report IP
      addresses both as integer array (as before) and as human readable
      string (new addition).

## Changes in systemd-boot and the stub:

    * The timeout in the boot menu can be configured with the
      io.systemd.boot.timeout= SMBIOS type 11 string.


    * A new LoaderEntryPreferred setting has been added to systemd-boot that
      is similar to LoaderEntryDefault, but that takes into the account boot
      assessment logic, and will skip entries that have the tries-left counter
      set to zero.


    * bootctl's Varlink interface gained a new Install() method for performing
      systemd-boot installation/upgrade via IPC calls.


    * bootctl gained a new --efi-boot-option-description-with-device=yes
      switch which augments the EFI boot option description registered with
      the firmware to include information about the disk used for
      booting. This is useful when installing multiple OSes on the same
      system, but on different disks. (Example: install a main OS on the
      SSD of a laptop, plus another one on an USB stick.)

Changes in sd-varlink:

    * The Varlink implementation now supports SD_VARLINK_ANY as a wildcard
      type. This is useful to declare generic interfaces which need to
      support multiple types.


    * When sd_varlink_connect_url() is invoked with an unrecognized URL
      scheme, but an executable named after the scheme exists under
      /usr/lib/systemd/varlink-bridges/, it is invoked and receives an
      AF_UNIX socktpair() via the usual $LISTEN_FDS socket activation
      protocol. The aforementioned Varlink HTTP bridge project makes use of
      this to allow any local Varlink client (including varlinkctl) to
      contact remote Varlink services via HTTP. The concept is entirely
      generic however, and can be used to plug in arbitrary other transport
      protocols, proxies, or connection setup mechanisms.

Changes in systemd-resolved:

    * systemd-resolved's .delegate files learnt a new setting FirewallMark=
      to set the Linux network stack's "firewall mark" value for all DNS
      traffic generated by the delegation.


    * resolvectl now uses Varlink to connect to systemd-resolved.


    * Queries done through nss-resolve can be limited to a specific
      interface with the $SYSTEMD_NSS_RESOLVE_INTERFACE environment
      variable.


    * systemd-resolved now supports ifindex=0 in the BrowseServices IPC API,
      to allow browsing all mDNS interfaces in one call.

## Changes in systemd-sysupdate/systemd-sysupdated:

    * systemd-sysupdate gained a new 'acquire' verb, allowing the download
      and installation or update steps to be done separately.


    * systemd-sysupdate will now refuse processing SHA256SUMS manifests if
      they list a file BEST-BEFORE- suffixed by a date that is already in
      the past, as a simple mechanism to detect freshness.


    * systemd-sysupdate now can mark partitions as partially downloaded.

Changes in systemd-vmspawn:

    * systemd-vmspawn gained support for registering with systemd-machined
      in the user session. New options --user/--system control which
      instance is used.


    * systemd-vmspawn gained support for ephemeral machines via a new
      --ephemeral option. This is similar to the functionality provided via
      the same switch in systemd-nspawn.


    * systemd-vmspawn gained a new switch --image-format= for selecting the
      image format (i.e. support qcow2 in additin to raw) to boot from.
      Also --extra-drive= now takes the image format as a colon separated
      parameter.

## Changes in systemd-nsresourced/systemd-mountfsd:

    * The MakeDirectory() Varlink IPC call provided by systemd-mountfsd now
      accepts a "mode" parameter for configuring the access mode of the
      newly created directory. The MountImage() call gained a new
      "mountOptions" parameter for configuring mount options for the
      various partitions of a DDI explicitly. The call will now also report
      via a new "singleFileSystem" field in the response whether it is
      processing a DDI lacking a GPT envelope, and consisting of a raw file
      system only. A new input parameter "relaxExtensionReleaseChecks"
      controls whether to enforce extension release checks.


    * systemd-nsresourced's BPF-LSM based security policy on user
      namespaces it delegates UID ranges too is relaxed: processes in such
      namespaces may now freely access to inodes owned by UIDs/GIDs outside
      of the transient UID range. This reflects the fact that the security
      policy exists to ensure ownership of inodes by transient UIDs is
      never persisted on disk.


    * systemd-nsresourced can now delegate multiples of additional 64K
      ranges of UIDs/GIDs to user namespaces, on request. This permits
      nesting of user namespace enabled containers with transient UID
      ranges.


    * systemd-nsresourced now supports a new type user namespace UID
      delegation: only the client's UID is mapped. This is very similar to
      what the kernel allows anyway as unprivileged delegation without
      systemd-nsresourced involvement, however, can be combined with
      multiple additional 64K ranges (see above).


    * systemd-nsresourced may now optionally map the "foreign" UID/GID
      range to itself for user namespaces it delegates transient UIDs/GIDs
      to. This opens up the concept for nested containers.


    * systemd-nsresourced's and systemd-mountfsd's Varlink sockets may now
      be mounted into container trees, to permit nested use of their
      functionality. This can be used automatically in systemd-nspawn's
      --private-users-delegate= option.

Changes in systemd-logind:

    * systemd-logind/systemd-udevd gained support for a new "xaccess"
      concept for delegating access to specific devices to users with
      specially marked sessions. The augments the "uaccess" logic that
      provides device access to users with foreground sessions. The primary
      usecase for this is to give access to GPU render devices to local
      graphical sessions for remote users, i.e. which are not attached to
      any local seat. Sessions are configured via the PAM environment
      variable XDG_SESSION_EXTRA_DEVICE_ACCESS= for this logic.


    * systemd-inhibit --list option gained support for JSON output
      and filtering with --what= , --who=, --why=, and --mode=.

Changes in systemd-portabled:

    * systemd-portabled now also runs as a user service. Unprivileged users
      can run portable services (on sufficiently fresh
      kernels). portablectl gained a pair of switches --user/--system to
      explicitly select which service instance to talk to.


    * systemd-portabled will now generate a policy and pin the image for a
      portable service, so that the image cannot be changed later without a
      reattach.

Changes in other components:

    * systemd-repart gained basic support for dm-integrity protection of
      encrypted volumes. Two new options Integrity= and IntegrityAlgorithm=
      can be used to configure integrity checks for LUKS volumes.


    * Image dissection policies have been extended to allow restricting
      file system types and requiring integrity checks for encrypted
      volumes with a new 'encryptedwithintegrity' policy.


    * systemd-dissect gained a --copy-ownership= switch to configure
      ownership of copied files.


    * systemd-keyutil gained an 'extract-certificate' verb to print the
      X.509 certificate. The existing 'public' verb has been renamed to
      'extract-public' as it works analogously. (The old name remains
      available for compatibility.)


    * Support for interactive polkit authorization has been added to
      systemd-sysext and varlinkctl.


    * A polkit policy was added for systemd-ask-password, allowing it to be
      used by unprivileged callers.


    * journalctl now implements a Varlink interface that exposes a
      GetEntries() method, which allows retrieving journal entries.


    * systemd-importd gained support for downloading OCI images ("importctl
      pull-oci"). They will be stored locally as "mstack" images, which
      then can be used by various components, for example be run as system
      services via RootMStack= in unit files, or as systemd-nspawn
      containers via --mstack= (see below).


    * systemd-nspawn gained a new --mstack= parameter to support the new
      "mstack" feature for containers.


    * A new systemd-mstack command line tool has been introduced to support
      the new "mstack" feature interactively.


    * New options SYSTEMD_COLORS=auto-16, SYSTEMD_COLORS=auto-256, and
      SYSTEMD_COLORS=auto-24bit have been added. They are like
      SYSTEMD_COLORS=16, SYSTEMD_COLORS=256, and SYSTEMD_COLORS=24bit
      respectively when output is to a non-dumb TTY, and degrade to
      SYSTEMD_COLORS=no otherwise.


    * Standalone versions of the systemd-sysusers and systemd-tmpfiles
      binaries now support full functionality. (Previously, those tools
      were compiled without features which would require libmount and pull
      in a lot of dependencies. After the conversion to dlopen the option
      whether to support features requiring libmount is made by providing
      libmount or not, at install time.)


    * Internal code dealing with processes has been updated to use pidfds
      in many places.


    * busctl's 'wait' verb now honours --limit-messages= too.


    * systemd-cryptsetup gained support for a new fixate-volume-key=
      option, that can be used to pin a specific encrypted volume to an
      /etc/crypttab entry via its volume key (more precisely a hash derived
      from it). systemd-repart will assist generating this information.


    * systemd-sysext/systemd-confext's "refresh" will now by default try to
      suppress any operation in case no images where added, removed or
      changed. To force a umount/mount operation in this case (i.e. get
      back to the status quo ante) a new --always-refresh= option has been
      added.


    * systemd-oomd acquired "prekill hook" functionality, permitting other
      system components to synchronously hook into the OOM killing logic,
      by registering a Varlink socket in a special directory.


    * systemd-analyze learnt a new verb "identify-tpm2" which shows
      vendor/model information extracted from the system's TPM.

## Changes in units:

    * runlevel[0-6].target units that were removed in v258 have been
      restored and can be enabled with the new -Dcompat-sysv-interfaces=yes
      meson option. The installation of legacy.conf for tmpfiles is now
      also conditionalized under the same option.


    * getty@.service gained an [Install] and must now be explicitly enabled
      to be active.

Contributors

    Contributions from: A S Alam, Adam Williamson, Adrian Vovk,
    Alessandro Astone, Alexis-Emmanuel Haeringer, Allison Karlitskaya,
    Américo Monteiro, Andrii Zora, André Paiusco, Anton Tiurin,
    Antonio Alvarez Feijoo, Arjun-C-S, Artur Kowalski,
    AshishKumar Mishra, Baurzhan Muftakhidinov, Ben Boeckel,
    Betacentury, Bouke van der Bijl, Carlos Peón Costa,
    Carolina Jubran, Cathy Hu, Chris Down, Chris Lindee,
    Christian Brauner, Christian Glombek, Christian Hesse,
    Christopher Cooper, Christopher Head,
    Copilot Autofix powered by AI, Cyrus Xi, Daan De Meyer,
    Dan McGregor, Daniel Foster, Daniel Nylander, Daniel Rusek,
    David Santamaría Rogado, David Tardon, Derek J. Clark, Dirk Su,
    Dmitry V. Levin, Dmytro Bagrii, Dylan M. Taylor,
    Efstathios Iosifidis, Eisuke Kawashima, Ettore Atalan, Fergus Dall,
    Florian Klink, Franck Bui, Frantisek Sumsal, Govind Venugopal,
    Graham Reed, Guiorgy, Han Sol Jin, Hans de Goede, Heran Yang,
    IntenseWiggling, Ivan Kruglov, Ivan Shapovalov, James Le Cuirot,
    Jan Kuparinen, Jeff Layton, Jeremy Kerr, Jesse Guo, Jian Wen,
    Jim Spentzos, Julian Sparber, Jörg Behrmann, Kai Lüke,
    Lennart Poettering, Louis Stagg, Luca Boccassi, Lucas Werkmeister,
    Luiz Amaral, Léane GRASSER, Malcolm Frazier, Marc Pervaz Boocha,
    Marcel Leismann, Mario Limonciello, Mario Limonciello (AMD),
    Martin Srebotnjak, Matt Fleming, Matteo Croce, Matthijs Kooijman,
    Max Gautier, Maximilian Bosch, Miao Wang, Michael Vogt,
    Michal Sekletár, Mike Gilbert, Mike Yuan, Mikhail Novosyolov,
    Nandakumar Raghavan, Nick Rosbrook, Nicolas Dorier, Oblivionsage,
    Oleksandr Andrushchenko, Oğuz Ersen, Pablo Fraile Alonso,
    Peter Oliver, Philip Withnall, Pontus Lundkvist, Popax21,
    Rito Rhymes, Rodrigo Campos, Ronan Pigott, Ryan Zeigler,
    Salvatore Cocuzza, Sergey A., Skye Soss, Sriman Achanta,
    Tabis Kabis, Temuri Doghonadze, The-An0nym, Thomas Weißschuh,
    Thorsten Kukuk, Tobias Heider, Tobias Stoeckmann, Ulrich Ölmann,
    Usama Arif, Val Markovic, Vitaly Kuznetsov, Vunny Sodhi,
    Weixie Cui, Yaping Li, Yaron Shahrabani, Yu Watanabe,
    Yuri Chornoivan, ZauberNerd, Zbigniew Jędrzejewski-Szmek,
    Zongyuan He, andre4ik3, calm329, cdown, cyclopentane, davidak,
    dongshengyuan, francescoza6, gvenugo3, joo es, kiamvdd, lumingzh,
    naly zzwd, nikstur, novenary, noxiouz, patrick, ppkramer-hub, r-vdp,
    safforddr, scarlet-storm, sd416, seidlerv, smosia, tuhaowen, zefr0x

    — Edinburgh, 2026/03/17