man/systemd-nspawn.xml

   1 <?xml version='1.0'?> <!--*- Mode: nxml; nxml-child-indent: 2; indent-tabs-mode: nil -*-->
   2 <!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN"
   3   "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd" [
   4 <!ENTITY fedora_latest_version "28">
   5 <!ENTITY fedora_cloud_release "1.1">
   6 ]>
   7
   8 <!--
   9   SPDX-License-Identifier: LGPL-2.1+
  10
  11   Copyright 2010 Lennart Poettering
  12 -->
  13
  14 <refentry id="systemd-nspawn"
  15     xmlns:xi="http://www.w3.org/2001/XInclude">
  16
  17   <refentryinfo>
  18     <title>systemd-nspawn</title>
  19     <productname>systemd</productname>
  20
  21     <authorgroup>
  22       <author>
  23         <contrib>Developer</contrib>
  24         <firstname>Lennart</firstname>
  25         <surname>Poettering</surname>
  26         <email>lennart@poettering.net</email>
  27       </author>
  28     </authorgroup>
  29   </refentryinfo>
  30
  31   <refmeta>
  32     <refentrytitle>systemd-nspawn</refentrytitle>
  33     <manvolnum>1</manvolnum>
  34   </refmeta>
  35
  36   <refnamediv>
  37     <refname>systemd-nspawn</refname>
  38     <refpurpose>Spawn a namespace container for debugging, testing and building</refpurpose>
  39   </refnamediv>
  40
  41   <refsynopsisdiv>
  42     <cmdsynopsis>
  43       <command>systemd-nspawn</command>
  44       <arg choice="opt" rep="repeat">OPTIONS</arg>
  45       <arg choice="opt"><replaceable>COMMAND</replaceable>
  46       <arg choice="opt" rep="repeat">ARGS</arg>
  47       </arg>
  48     </cmdsynopsis>
  49     <cmdsynopsis>
  50       <command>systemd-nspawn</command>
  51       <arg choice="plain">--boot</arg>
  52       <arg choice="opt" rep="repeat">OPTIONS</arg>
  53       <arg choice="opt" rep="repeat">ARGS</arg>
  54     </cmdsynopsis>
  55   </refsynopsisdiv>
  56
  57   <refsect1>
  58     <title>Description</title>
  59
  60     <para><command>systemd-nspawn</command> may be used to run a command or OS in a light-weight namespace
  61     container. In many ways it is similar to <citerefentry
  62     project='man-pages'><refentrytitle>chroot</refentrytitle><manvolnum>1</manvolnum></citerefentry>, but more powerful
  63     since it fully virtualizes the file system hierarchy, as well as the process tree, the various IPC subsystems and
  64     the host and domain name.</para>
  65
  66     <para><command>systemd-nspawn</command> may be invoked on any directory tree containing an operating system tree,
  67     using the <option>--directory=</option> command line option. By using the <option>--machine=</option> option an OS
  68     tree is automatically searched for in a couple of locations, most importantly in
  69     <filename>/var/lib/machines</filename>, the suggested directory to place container images installed on the
  70     system.</para>
  71
  72     <para>In contrast to <citerefentry
  73     project='man-pages'><refentrytitle>chroot</refentrytitle><manvolnum>1</manvolnum></citerefentry> <command>systemd-nspawn</command>
  74     may be used to boot full Linux-based operating systems in a container.</para>
  75
  76     <para><command>systemd-nspawn</command> limits access to various kernel interfaces in the container to read-only,
  77     such as <filename>/sys</filename>, <filename>/proc/sys</filename> or <filename>/sys/fs/selinux</filename>. The
  78     host's network interfaces and the system clock may not be changed from within the container. Device nodes may not
  79     be created. The host system cannot be rebooted and kernel modules may not be loaded from within the
  80     container.</para>
  81
  82     <para>Use a tool like <citerefentry
  83     project='mankier'><refentrytitle>dnf</refentrytitle><manvolnum>8</manvolnum></citerefentry>, <citerefentry
  84     project='die-net'><refentrytitle>debootstrap</refentrytitle><manvolnum>8</manvolnum></citerefentry>, or
  85     <citerefentry project='archlinux'><refentrytitle>pacman</refentrytitle><manvolnum>8</manvolnum></citerefentry> to
  86     set up an OS directory tree suitable as file system hierarchy for <command>systemd-nspawn</command> containers. See
  87     the Examples section below for details on suitable invocation of these commands.</para>
  88
  89     <para>As a safety check <command>systemd-nspawn</command> will verify the existence of
  90     <filename>/usr/lib/os-release</filename> or <filename>/etc/os-release</filename> in the container tree before
  91     starting the container (see
  92     <citerefentry><refentrytitle>os-release</refentrytitle><manvolnum>5</manvolnum></citerefentry>).  It might be
  93     necessary to add this file to the container tree manually if the OS of the container is too old to contain this
  94     file out-of-the-box.</para>
  95
  96     <para><command>systemd-nspawn</command> may be invoked directly from the interactive command line or run as system
  97     service in the background. In this mode each container instance runs as its own service instance; a default
  98     template unit file <filename>systemd-nspawn@.service</filename> is provided to make this easy, taking the container
  99     name as instance identifier. Note that different default options apply when <command>systemd-nspawn</command> is
 100     invoked by the template unit file than interactively on the command line. Most importantly the template unit file
 101     makes use of the <option>--boot</option> which is not the default in case <command>systemd-nspawn</command> is
 102     invoked from the interactive command line. Further differences with the defaults are documented along with the
 103     various supported options below.</para>
 104
 105     <para>The <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry> tool may
 106     be used to execute a number of operations on containers. In particular it provides easy-to-use commands to run
 107     containers as system services using the <filename>systemd-nspawn@.service</filename> template unit
 108     file.</para>
 109
 110     <para>Along with each container a settings file with the <filename>.nspawn</filename> suffix may exist, containing
 111     additional settings to apply when running the container. See
 112     <citerefentry><refentrytitle>systemd.nspawn</refentrytitle><manvolnum>5</manvolnum></citerefentry> for
 113     details. Settings files override the default options used by the <filename>systemd-nspawn@.service</filename>
 114     template unit file, making it usually unnecessary to alter this template file directly.</para>
 115
 116     <para>Note that <command>systemd-nspawn</command> will mount file systems private to the container to
 117     <filename>/dev</filename>, <filename>/run</filename> and similar. These will not be visible outside of the
 118     container, and their contents will be lost when the container exits.</para>
 119
 120     <para>Note that running two <command>systemd-nspawn</command> containers from the same directory tree will not make
 121     processes in them see each other. The PID namespace separation of the two containers is complete and the containers
 122     will share very few runtime objects except for the underlying file system. Use
 123     <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry>'s
 124     <command>login</command> or <command>shell</command> commands to request an additional login session in a running
 125     container.</para>
 126
 127     <para><command>systemd-nspawn</command> implements the <ulink
 128     url="https://www.freedesktop.org/wiki/Software/systemd/ContainerInterface">Container Interface</ulink>
 129     specification.</para>
 130
 131     <para>While running, containers invoked with <command>systemd-nspawn</command> are registered with the
 132     <citerefentry><refentrytitle>systemd-machined</refentrytitle><manvolnum>8</manvolnum></citerefentry> service that
 133     keeps track of running containers, and provides programming interfaces to interact with them.</para>
 134   </refsect1>
 135
 136   <refsect1>
 137     <title>Options</title>
 138
 139     <para>If option <option>-b</option> is specified, the arguments
 140     are used as arguments for the init program. Otherwise,
 141     <replaceable>COMMAND</replaceable> specifies the program to launch
 142     in the container, and the remaining arguments are used as
 143     arguments for this program. If <option>--boot</option> is not used and
 144     no arguments are specified, a shell is launched in the
 145     container.</para>
 146
 147     <para>The following options are understood:</para>
 148
 149     <variablelist>
 150       <varlistentry>
 151         <term><option>-D</option></term>
 152         <term><option>--directory=</option></term>
 153
 154         <listitem><para>Directory to use as file system root for the
 155         container.</para>
 156
 157         <para>If neither <option>--directory=</option>, nor
 158         <option>--image=</option> is specified the directory is
 159         determined by searching for a directory named the same as the
 160         machine name specified with <option>--machine=</option>. See
 161         <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry>
 162         section "Files and Directories" for the precise search path.</para>
 163
 164         <para>If neither <option>--directory=</option>,
 165         <option>--image=</option>, nor <option>--machine=</option>
 166         are specified, the current directory will
 167         be used. May not be specified together with
 168         <option>--image=</option>.</para></listitem>
 169       </varlistentry>
 170
 171       <varlistentry>
 172         <term><option>--template=</option></term>
 173
 174         <listitem><para>Directory or <literal>btrfs</literal> subvolume to use as template for the container's root
 175         directory. If this is specified and the container's root directory (as configured by
 176         <option>--directory=</option>) does not yet exist it is created as <literal>btrfs</literal> snapshot (if
 177         supported) or plain directory (otherwise) and populated from this template tree. Ideally, the specified
 178         template path refers to the root of a <literal>btrfs</literal> subvolume, in which case a simple copy-on-write
 179         snapshot is taken, and populating the root directory is instant. If the specified template path does not refer
 180         to the root of a <literal>btrfs</literal> subvolume (or not even to a <literal>btrfs</literal> file system at
 181         all), the tree is copied (though possibly in a copy-on-write scheme — if the file system supports that), which
 182         can be substantially more time-consuming. May not be specified together with <option>--image=</option> or
 183         <option>--ephemeral</option>.</para>
 184
 185         <para>Note that this switch leaves host name, machine ID and
 186         all other settings that could identify the instance
 187         unmodified.</para></listitem>
 188       </varlistentry>
 189
 190       <varlistentry>
 191         <term><option>-x</option></term>
 192         <term><option>--ephemeral</option></term>
 193
 194         <listitem><para>If specified, the container is run with a temporary snapshot of its file system that is removed
 195         immediately when the container terminates. May not be specified together with
 196         <option>--template=</option>.</para>
 197         <para>Note that this switch leaves host name, machine ID and
 198         all other settings that could identify the instance
 199         unmodified.</para></listitem>
 200       </varlistentry>
 201
 202       <varlistentry>
 203         <term><option>-i</option></term>
 204         <term><option>--image=</option></term>
 205
 206         <listitem><para>Disk image to mount the root directory for the
 207         container from. Takes a path to a regular file or to a block
 208         device node. The file or block device must contain
 209         either:</para>
 210
 211         <itemizedlist>
 212           <listitem><para>An MBR partition table with a single
 213           partition of type 0x83 that is marked
 214           bootable.</para></listitem>
 215
 216           <listitem><para>A GUID partition table (GPT) with a single
 217           partition of type
 218           0fc63daf-8483-4772-8e79-3d69d8477de4.</para></listitem>
 219
 220           <listitem><para>A GUID partition table (GPT) with a marked
 221           root partition which is mounted as the root directory of the
 222           container. Optionally, GPT images may contain a home and/or
 223           a server data partition which are mounted to the appropriate
 224           places in the container. All these partitions must be
 225           identified by the partition types defined by the <ulink
 226           url="https://www.freedesktop.org/wiki/Specifications/DiscoverablePartitionsSpec/">Discoverable
 227           Partitions Specification</ulink>.</para></listitem>
 228
 229           <listitem><para>No partition table, and a single file system spanning the whole image.</para></listitem>
 230         </itemizedlist>
 231
 232         <para>On GPT images, if an EFI System Partition (ESP) is discovered, it is automatically mounted to
 233         <filename>/efi</filename> (or <filename>/boot</filename> as fallback) in case a directory by this name exists
 234         and is empty.</para>
 235
 236         <para>Partitions encrypted with LUKS are automatically decrypted. Also, on GPT images dm-verity data integrity
 237         hash partitions are set up if the root hash for them is specified using the <option>--root-hash=</option>
 238         option.</para>
 239
 240         <para>Any other partitions, such as foreign partitions or swap partitions are not mounted. May not be specified
 241         together with <option>--directory=</option>, <option>--template=</option>.</para></listitem>
 242       </varlistentry>
 243
 244       <varlistentry>
 245         <term><option>--root-hash=</option></term>
 246
 247         <listitem><para>Takes a data integrity (dm-verity) root hash specified in hexadecimal. This option enables data
 248         integrity checks using dm-verity, if the used image contains the appropriate integrity data (see above). The
 249         specified hash must match the root hash of integrity data, and is usually at least 256 bits (and hence 64
 250         formatted hexadecimal characters) long (in case of SHA256 for example). If this option is not specified, but
 251         the image file carries the <literal>user.verity.roothash</literal> extended file attribute (see <citerefentry
 252         project='man-pages'><refentrytitle>xattr</refentrytitle><manvolnum>7</manvolnum></citerefentry>), then the root
 253         hash is read from it, also as formatted hexadecimal characters. If the extended file attribute is not found (or
 254         is not supported by the underlying file system), but a file with the <filename>.roothash</filename> suffix is
 255         found next to the image file, bearing otherwise the same name, the root hash is read from it and automatically
 256         used, also as formatted hexadecimal characters.</para></listitem>
 257       </varlistentry>
 258
 259       <varlistentry>
 260         <term><option>-a</option></term>
 261         <term><option>--as-pid2</option></term>
 262
 263         <listitem><para>Invoke the shell or specified program as process ID (PID) 2 instead of PID 1 (init). By
 264         default, if neither this option nor <option>--boot</option> is used, the selected program is run as the process
 265         with PID 1, a mode only suitable for programs that are aware of the special semantics that the process with
 266         PID 1 has on UNIX. For example, it needs to reap all processes reparented to it, and should implement
 267         <command>sysvinit</command> compatible signal handling (specifically: it needs to reboot on SIGINT, reexecute
 268         on SIGTERM, reload configuration on SIGHUP, and so on). With <option>--as-pid2</option> a minimal stub init
 269         process is run as PID 1 and the selected program is executed as PID 2 (and hence does not need to implement any
 270         special semantics). The stub init process will reap processes as necessary and react appropriately to
 271         signals. It is recommended to use this mode to invoke arbitrary commands in containers, unless they have been
 272         modified to run correctly as PID 1. Or in other words: this switch should be used for pretty much all commands,
 273         except when the command refers to an init or shell implementation, as these are generally capable of running
 274         correctly as PID 1. This option may not be combined with <option>--boot</option>.</para>
 275         </listitem>
 276       </varlistentry>
 277
 278       <varlistentry>
 279         <term><option>-b</option></term>
 280         <term><option>--boot</option></term>
 281
 282         <listitem><para>Automatically search for an init program and invoke it as PID 1, instead of a shell or a user
 283         supplied program. If this option is used, arguments specified on the command line are used as arguments for the
 284         init program. This option may not be combined with <option>--as-pid2</option>.</para>
 285
 286         <para>The following table explains the different modes of invocation and relationship to
 287         <option>--as-pid2</option> (see above):</para>
 288
 289         <table>
 290           <title>Invocation Mode</title>
 291           <tgroup cols='2' align='left' colsep='1' rowsep='1'>
 292             <colspec colname="switch" />
 293             <colspec colname="explanation" />
 294             <thead>
 295               <row>
 296                 <entry>Switch</entry>
 297                 <entry>Explanation</entry>
 298               </row>
 299             </thead>
 300             <tbody>
 301               <row>
 302                 <entry>Neither <option>--as-pid2</option> nor <option>--boot</option> specified</entry>
 303                 <entry>The passed parameters are interpreted as the command line, which is executed as PID 1 in the container.</entry>
 304               </row>
 305
 306               <row>
 307                 <entry><option>--as-pid2</option> specified</entry>
 308                 <entry>The passed parameters are interpreted as the command line, which is executed as PID 2 in the container. A stub init process is run as PID 1.</entry>
 309               </row>
 310
 311               <row>
 312                 <entry><option>--boot</option> specified</entry>
 313                 <entry>An init program is automatically searched for and run as PID 1 in the container. The passed parameters are used as invocation parameters for this process.</entry>
 314               </row>
 315
 316             </tbody>
 317           </tgroup>
 318         </table>
 319
 320         <para>Note that <option>--boot</option> is the default mode of operation if the
 321         <filename>systemd-nspawn@.service</filename> template unit file is used.</para>
 322         </listitem>
 323       </varlistentry>
 324
 325       <varlistentry>
 326         <term><option>--chdir=</option></term>
 327
 328         <listitem><para>Change to the specified working directory before invoking the process in the container. Expects
 329         an absolute path in the container's file system namespace.</para></listitem>
 330       </varlistentry>
 331
 332       <varlistentry>
 333         <term><option>--pivot-root=</option></term>
 334
 335         <listitem><para>Pivot the specified directory to <filename>/</filename> inside the container, and either unmount the
 336         container's old root, or pivot it to another specified directory. Takes one of: a path argument — in which case the
 337         specified path will be pivoted to <filename>/</filename> and the old root will be unmounted; or a colon-separated pair
 338         of new root path and pivot destination for the old root. The new root path will be pivoted to <filename>/</filename>,
 339         and the old <filename>/</filename> will be pivoted to the other directory. Both paths must be absolute, and are resolved
 340         in the container's file system namespace.</para>
 341
 342         <para>This is for containers which have several bootable directories in them; for example, several
 343         <ulink url="https://ostree.readthedocs.io/en/latest/">OSTree</ulink> deployments. It emulates the behavior of
 344         the boot loader and initial RAM disk which normally select which directory to mount as the root and start the
 345         container's PID 1 in.</para></listitem>
 346       </varlistentry>
 347
 348       <varlistentry>
 349         <term><option>-u</option></term>
 350         <term><option>--user=</option></term>
 351
 352         <listitem><para>After transitioning into the container, change
 353         to the specified user-defined in the container's user
 354         database. Like all other systemd-nspawn features, this is not
 355         a security feature and provides protection against accidental
 356         destructive operations only.</para></listitem>
 357       </varlistentry>
 358
 359       <varlistentry>
 360         <term><option>-M</option></term>
 361         <term><option>--machine=</option></term>
 362
 363         <listitem><para>Sets the machine name for this container. This
 364         name may be used to identify this container during its runtime
 365         (for example in tools like
 366         <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry>
 367         and similar), and is used to initialize the container's
 368         hostname (which the container can choose to override,
 369         however). If not specified, the last component of the root
 370         directory path of the container is used, possibly suffixed
 371         with a random identifier in case <option>--ephemeral</option>
 372         mode is selected. If the root directory selected is the host's
 373         root directory the host's hostname is used as default
 374         instead.</para></listitem>
 375       </varlistentry>
 376
 377       <varlistentry>
 378         <term><option>--hostname=</option></term>
 379
 380         <listitem><para>Controls the hostname to set within the container, if different from the machine name. Expects
 381         a valid hostname as argument. If this option is used, the kernel hostname of the container will be set to this
 382         value, otherwise it will be initialized to the machine name as controlled by the <option>--machine=</option>
 383         option described above. The machine name is used for various aspect of identification of the container from the
 384         outside, the kernel hostname configurable with this option is useful for the container to identify itself from
 385         the inside. It is usually a good idea to keep both forms of identification synchronized, in order to avoid
 386         confusion. It is hence recommended to avoid usage of this option, and use <option>--machine=</option>
 387         exclusively. Note that regardless whether the container's hostname is initialized from the name set with
 388         <option>--hostname=</option> or the one set with <option>--machine=</option>, the container can later override
 389         its kernel hostname freely on its own as well.</para>
 390         </listitem>
 391       </varlistentry>
 392
 393       <varlistentry>
 394         <term><option>--uuid=</option></term>
 395
 396         <listitem><para>Set the specified UUID for the container. The
 397         init system will initialize
 398         <filename>/etc/machine-id</filename> from this if this file is
 399         not set yet. Note that this option takes effect only if
 400         <filename>/etc/machine-id</filename> in the container is
 401         unpopulated.</para></listitem>
 402       </varlistentry>
 403
 404       <varlistentry>
 405         <term><option>-S</option></term>
 406         <term><option>--slice=</option></term>
 407
 408         <listitem><para>Make the container part of the specified slice, instead of the default
 409         <filename>machine.slice</filename>. This applies only if the machine is run in its own scope unit, i.e. if
 410         <option>--keep-unit</option> isn't used.</para>
 411         </listitem>
 412       </varlistentry>
 413
 414       <varlistentry>
 415         <term><option>--property=</option></term>
 416
 417         <listitem><para>Set a unit property on the scope unit to register for the machine. This applies only if the
 418         machine is run in its own scope unit, i.e. if <option>--keep-unit</option> isn't used. Takes unit property
 419         assignments in the same format as <command>systemctl set-property</command>. This is useful to set memory
 420         limits and similar for container.</para>
 421         </listitem>
 422       </varlistentry>
 423
 424       <varlistentry>
 425         <term><option>--private-users=</option></term>
 426
 427         <listitem><para>Controls user namespacing. If enabled, the container will run with its own private set of UNIX
 428         user and group ids (UIDs and GIDs). This involves mapping the private UIDs/GIDs used in the container (starting
 429         with the container's root user 0 and up) to a range of UIDs/GIDs on the host that are not used for other
 430         purposes (usually in the range beyond the host's UID/GID 65536). The parameter may be specified as follows:</para>
 431
 432         <orderedlist>
 433           <listitem><para>If one or two colon-separated numbers are specified, user namespacing is turned on. The first
 434           parameter specifies the first host UID/GID to assign to the container, the second parameter specifies the
 435           number of host UIDs/GIDs to assign to the container. If the second parameter is omitted, 65536 UIDs/GIDs are
 436           assigned.</para></listitem>
 437
 438           <listitem><para>If the parameter is omitted, or true, user namespacing is turned on. The UID/GID range to
 439           use is determined automatically from the file ownership of the root directory of the container's directory
 440           tree. To use this option, make sure to prepare the directory tree in advance, and ensure that all files and
 441           directories in it are owned by UIDs/GIDs in the range you'd like to use. Also, make sure that used file ACLs
 442           exclusively reference UIDs/GIDs in the appropriate range. If this mode is used the number of UIDs/GIDs
 443           assigned to the container for use is 65536, and the UID/GID of the root directory must be a multiple of
 444           65536.</para></listitem>
 445
 446           <listitem><para>If the parameter is false, user namespacing is turned off. This is the default.</para>
 447           </listitem>
 448
 449           <listitem><para>The special value <literal>pick</literal> turns on user namespacing. In this case the UID/GID
 450           range is automatically chosen. As first step, the file owner of the root directory of the container's
 451           directory tree is read, and it is checked that it is currently not used by the system otherwise (in
 452           particular, that no other container is using it). If this check is successful, the UID/GID range determined
 453           this way is used, similar to the behavior if "yes" is specified. If the check is not successful (and thus
 454           the UID/GID range indicated in the root directory's file owner is already used elsewhere) a new – currently
 455           unused – UID/GID range of 65536 UIDs/GIDs is randomly chosen between the host UID/GIDs of 524288 and
 456           1878982656, always starting at a multiple of 65536. This setting implies
 457           <option>--private-users-chown</option> (see below), which has the effect that the files and directories in
 458           the container's directory tree will be owned by the appropriate users of the range picked. Using this option
 459           makes user namespace behavior fully automatic. Note that the first invocation of a previously unused
 460           container image might result in picking a new UID/GID range for it, and thus in the (possibly expensive) file
 461           ownership adjustment operation. However, subsequent invocations of the container will be cheap (unless of
 462           course the picked UID/GID range is assigned to a different use by then).</para></listitem>
 463         </orderedlist>
 464
 465         <para>It is recommended to assign at least 65536 UIDs/GIDs to each container, so that the usable UID/GID range in the
 466         container covers 16 bit. For best security, do not assign overlapping UID/GID ranges to multiple containers. It is
 467         hence a good idea to use the upper 16 bit of the host 32-bit UIDs/GIDs as container identifier, while the lower 16
 468         bit encode the container UID/GID used. This is in fact the behavior enforced by the
 469         <option>--private-users=pick</option> option.</para>
 470
 471         <para>When user namespaces are used, the GID range assigned to each container is always chosen identical to the
 472         UID range.</para>
 473
 474         <para>In most cases, using <option>--private-users=pick</option> is the recommended option as it enhances
 475         container security massively and operates fully automatically in most cases.</para>
 476
 477         <para>Note that the picked UID/GID range is not written to <filename>/etc/passwd</filename> or
 478         <filename>/etc/group</filename>. In fact, the allocation of the range is not stored persistently anywhere,
 479         except in the file ownership of the files and directories of the container.</para>
 480
 481         <para>Note that when user namespacing is used file ownership on disk reflects this, and all of the container's
 482         files and directories are owned by the container's effective user and group IDs. This means that copying files
 483         from and to the container image requires correction of the numeric UID/GID values, according to the UID/GID
 484         shift applied.</para></listitem>
 485       </varlistentry>
 486
 487       <varlistentry>
 488         <term><option>--private-users-chown</option></term>
 489
 490         <listitem><para>If specified, all files and directories in the container's directory tree will adjusted so that
 491         they are owned to the appropriate UIDs/GIDs selected for the container (see above). This operation is
 492         potentially expensive, as it involves descending and iterating through the full directory tree of the
 493         container. Besides actual file ownership, file ACLs are adjusted as well.</para>
 494
 495         <para>This option is implied if <option>--private-users=pick</option> is used. This option has no effect if
 496         user namespacing is not used.</para></listitem>
 497       </varlistentry>
 498
 499       <varlistentry>
 500         <term><option>-U</option></term>
 501
 502         <listitem><para>If the kernel supports the user namespaces feature, equivalent to
 503         <option>--private-users=pick --private-users-chown</option>, otherwise equivalent to
 504         <option>--private-users=no</option>.</para>
 505
 506         <para>Note that <option>-U</option> is the default if the
 507         <filename>systemd-nspawn@.service</filename> template unit file is used.</para>
 508
 509         <para>Note: it is possible to undo the effect of <option>--private-users-chown</option> (or
 510         <option>-U</option>) on the file system by redoing the operation with the first UID of 0:</para>
 511
 512         <programlisting>systemd-nspawn … --private-users=0 --private-users-chown</programlisting>
 513         </listitem>
 514       </varlistentry>
 515
 516       <varlistentry>
 517         <term><option>--private-network</option></term>
 518
 519         <listitem><para>Disconnect networking of the container from
 520         the host. This makes all network interfaces unavailable in the
 521         container, with the exception of the loopback device and those
 522         specified with <option>--network-interface=</option> and
 523         configured with <option>--network-veth</option>. If this
 524         option is specified, the CAP_NET_ADMIN capability will be
 525         added to the set of capabilities the container retains. The
 526         latter may be disabled by using <option>--drop-capability=</option>.
 527         If this option is not specified (or implied by one of the options
 528         listed below), the container will have full access to the host network.
 529         </para></listitem>
 530       </varlistentry>
 531
 532       <varlistentry>
 533         <term><option>--network-namespace-path=</option></term>
 534
 535         <listitem><para>Takes the path to a file representing a kernel
 536         network namespace that the container shall run in. The specified path
 537         should refer to a (possibly bind-mounted) network namespace file, as
 538         exposed by the kernel below <filename>/proc/$PID/ns/net</filename>.
 539         This makes the container enter the given network namespace. One of the
 540         typical use cases is to give a network namespace under
 541         <filename>/run/netns</filename> created by <citerefentry
 542         project='man-pages'><refentrytitle>ip-netns</refentrytitle><manvolnum>8</manvolnum></citerefentry>,
 543         for example, <option>--network-namespace-path=/run/netns/foo</option>.
 544         Note that this option cannot be used together with other
 545         network-related options, such as <option>--private-network</option>
 546         or <option>--network-interface=</option>.</para></listitem>
 547       </varlistentry>
 548
 549       <varlistentry>
 550         <term><option>--network-interface=</option></term>
 551
 552         <listitem><para>Assign the specified network interface to the
 553         container. This will remove the specified interface from the
 554         calling namespace and place it in the container. When the
 555         container terminates, it is moved back to the host namespace.
 556         Note that <option>--network-interface=</option> implies
 557         <option>--private-network</option>. This option may be used
 558         more than once to add multiple network interfaces to the
 559         container.</para></listitem>
 560       </varlistentry>
 561
 562       <varlistentry>
 563         <term><option>--network-macvlan=</option></term>
 564
 565         <listitem><para>Create a <literal>macvlan</literal> interface
 566         of the specified Ethernet network interface and add it to the
 567         container. A <literal>macvlan</literal> interface is a virtual
 568         interface that adds a second MAC address to an existing
 569         physical Ethernet link. The interface in the container will be
 570         named after the interface on the host, prefixed with
 571         <literal>mv-</literal>. Note that
 572         <option>--network-macvlan=</option> implies
 573         <option>--private-network</option>. This option may be used
 574         more than once to add multiple network interfaces to the
 575         container.</para></listitem>
 576       </varlistentry>
 577
 578       <varlistentry>
 579         <term><option>--network-ipvlan=</option></term>
 580
 581         <listitem><para>Create an <literal>ipvlan</literal> interface
 582         of the specified Ethernet network interface and add it to the
 583         container. An <literal>ipvlan</literal> interface is a virtual
 584         interface, similar to a <literal>macvlan</literal> interface,
 585         which uses the same MAC address as the underlying interface.
 586         The interface in the container will be named after the
 587         interface on the host, prefixed with <literal>iv-</literal>.
 588         Note that <option>--network-ipvlan=</option> implies
 589         <option>--private-network</option>. This option may be used
 590         more than once to add multiple network interfaces to the
 591         container.</para></listitem>
 592       </varlistentry>
 593
 594       <varlistentry>
 595         <term><option>-n</option></term>
 596         <term><option>--network-veth</option></term>
 597
 598         <listitem><para>Create a virtual Ethernet link (<literal>veth</literal>) between host and container. The host
 599         side of the Ethernet link will be available as a network interface named after the container's name (as
 600         specified with <option>--machine=</option>), prefixed with <literal>ve-</literal>. The container side of the
 601         Ethernet link will be named <literal>host0</literal>. The <option>--network-veth</option> option implies
 602         <option>--private-network</option>.</para>
 603
 604         <para>Note that
 605         <citerefentry><refentrytitle>systemd-networkd.service</refentrytitle><manvolnum>8</manvolnum></citerefentry>
 606         includes by default a network file <filename>/usr/lib/systemd/network/80-container-ve.network</filename>
 607         matching the host-side interfaces created this way, which contains settings to enable automatic address
 608         provisioning on the created virtual link via DHCP, as well as automatic IP routing onto the host's external
 609         network interfaces. It also contains <filename>/usr/lib/systemd/network/80-container-host0.network</filename>
 610         matching the container-side interface created this way, containing settings to enable client side address
 611         assignment via DHCP. In case <filename>systemd-networkd</filename> is running on both the host and inside the
 612         container, automatic IP communication from the container to the host is thus available, with further
 613         connectivity to the external network.</para>
 614
 615         <para>Note that <option>--network-veth</option> is the default if the
 616         <filename>systemd-nspawn@.service</filename> template unit file is used.</para>
 617         </listitem>
 618       </varlistentry>
 619
 620       <varlistentry>
 621         <term><option>--network-veth-extra=</option></term>
 622
 623         <listitem><para>Adds an additional virtual Ethernet link
 624         between host and container. Takes a colon-separated pair of
 625         host interface name and container interface name. The latter
 626         may be omitted in which case the container and host sides will
 627         be assigned the same name. This switch is independent of
 628         <option>--network-veth</option>, and — in contrast — may be
 629         used multiple times, and allows configuration of the network
 630         interface names. Note that <option>--network-bridge=</option>
 631         has no effect on interfaces created with
 632         <option>--network-veth-extra=</option>.</para></listitem>
 633       </varlistentry>
 634
 635       <varlistentry>
 636         <term><option>--network-bridge=</option></term>
 637
 638         <listitem><para>Adds the host side of the Ethernet link created with <option>--network-veth</option> to the
 639         specified Ethernet bridge interface. Expects a valid network interface name of a bridge device as
 640         argument. Note that <option>--network-bridge=</option> implies <option>--network-veth</option>. If this option
 641         is used, the host side of the Ethernet link will use the <literal>vb-</literal> prefix instead of
 642         <literal>ve-</literal>.</para></listitem>
 643       </varlistentry>
 644
 645       <varlistentry>
 646         <term><option>--network-zone=</option></term>
 647
 648         <listitem><para>Creates a virtual Ethernet link (<literal>veth</literal>) to the container and adds it to an
 649         automatically managed Ethernet bridge interface. The bridge interface is named after the passed argument,
 650         prefixed with <literal>vz-</literal>. The bridge interface is automatically created when the first container
 651         configured for its name is started, and is automatically removed when the last container configured for its
 652         name exits. Hence, each bridge interface configured this way exists only as long as there's at least one
 653         container referencing it running. This option is very similar to <option>--network-bridge=</option>, besides
 654         this automatic creation/removal of the bridge device.</para>
 655
 656         <para>This setting makes it easy to place multiple related containers on a common, virtual Ethernet-based
 657         broadcast domain, here called a "zone". Each container may only be part of one zone, but each zone may contain
 658         any number of containers. Each zone is referenced by its name. Names may be chosen freely (as long as they form
 659         valid network interface names when prefixed with <literal>vz-</literal>), and it is sufficient to pass the same
 660         name to the <option>--network-zone=</option> switch of the various concurrently running containers to join
 661         them in one zone.</para>
 662
 663         <para>Note that
 664         <citerefentry><refentrytitle>systemd-networkd.service</refentrytitle><manvolnum>8</manvolnum></citerefentry>
 665         includes by default a network file <filename>/usr/lib/systemd/network/80-container-vz.network</filename>
 666         matching the bridge interfaces created this way, which contains settings to enable automatic address
 667         provisioning on the created virtual network via DHCP, as well as automatic IP routing onto the host's external
 668         network interfaces. Using <option>--network-zone=</option> is hence in most cases fully automatic and
 669         sufficient to connect multiple local containers in a joined broadcast domain to the host, with further
 670         connectivity to the external network.</para>
 671         </listitem>
 672       </varlistentry>
 673
 674       <varlistentry>
 675         <term><option>-p</option></term>
 676         <term><option>--port=</option></term>
 677
 678         <listitem><para>If private networking is enabled, maps an IP
 679         port on the host onto an IP port on the container. Takes a
 680         protocol specifier (either <literal>tcp</literal> or
 681         <literal>udp</literal>), separated by a colon from a host port
 682         number in the range 1 to 65535, separated by a colon from a
 683         container port number in the range from 1 to 65535. The
 684         protocol specifier and its separating colon may be omitted, in
 685         which case <literal>tcp</literal> is assumed. The container
 686         port number and its colon may be omitted, in which case the
 687         same port as the host port is implied. This option is only
 688         supported if private networking is used, such as with
 689         <option>--network-veth</option>, <option>--network-zone=</option>
 690         <option>--network-bridge=</option>.</para></listitem>
 691       </varlistentry>
 692
 693       <varlistentry>
 694         <term><option>-Z</option></term>
 695         <term><option>--selinux-context=</option></term>
 696
 697         <listitem><para>Sets the SELinux security context to be used
 698         to label processes in the container.</para>
 699         </listitem>
 700       </varlistentry>
 701
 702       <varlistentry>
 703         <term><option>-L</option></term>
 704         <term><option>--selinux-apifs-context=</option></term>
 705
 706         <listitem><para>Sets the SELinux security context to be used
 707         to label files in the virtual API file systems in the
 708         container.</para>
 709         </listitem>
 710       </varlistentry>
 711
 712       <varlistentry>
 713         <term><option>--capability=</option></term>
 714
 715         <listitem><para>List one or more additional capabilities to grant the container.
 716         Takes a comma-separated list of capability names, see
 717         <citerefentry project='man-pages'><refentrytitle>capabilities</refentrytitle><manvolnum>7</manvolnum></citerefentry>
 718         for more information. Note that the following capabilities will be granted in any way:
 719         CAP_AUDIT_CONTROL, CAP_AUDIT_WRITE, CAP_CHOWN, CAP_DAC_OVERRIDE, CAP_DAC_READ_SEARCH,
 720         CAP_FOWNER, CAP_FSETID, CAP_IPC_OWNER, CAP_KILL, CAP_LEASE, CAP_LINUX_IMMUTABLE,
 721         CAP_MKNOD, CAP_NET_BIND_SERVICE, CAP_NET_BROADCAST, CAP_NET_RAW, CAP_SETFCAP,
 722         CAP_SETGID, CAP_SETPCAP, CAP_SETUID, CAP_SYS_ADMIN, CAP_SYS_BOOT, CAP_SYS_CHROOT,
 723         CAP_SYS_NICE, CAP_SYS_PTRACE, CAP_SYS_RESOURCE, CAP_SYS_TTY_CONFIG. Also CAP_NET_ADMIN
 724         is retained if <option>--private-network</option> is specified.  If the special value
 725         <literal>all</literal> is passed, all capabilities are retained.</para></listitem>
 726       </varlistentry>
 727
 728       <varlistentry>
 729         <term><option>--drop-capability=</option></term>
 730
 731         <listitem><para>Specify one or more additional capabilities to
 732         drop for the container. This allows running the container with
 733         fewer capabilities than the default (see
 734         above).</para></listitem>
 735       </varlistentry>
 736
 737       <varlistentry>
 738         <term><option>--no-new-privileges=</option></term>
 739
 740         <listitem><para>Takes a boolean argument. Specifies the value of the <constant>PR_SET_NO_NEW_PRIVS</constant>
 741         flag for the container payload. Defaults to off. When turned on the payload code of the container cannot
 742         acquire new privileges, i.e. the "setuid" file bit as well as file system capabilities will not have an effect
 743         anymore. See <citerefentry
 744         project='man-pages'><refentrytitle>prctl</refentrytitle><manvolnum>2</manvolnum></citerefentry> for details
 745         about this flag. </para></listitem>
 746       </varlistentry>
 747
 748       <varlistentry>
 749         <term><option>--system-call-filter=</option></term>
 750
 751         <listitem><para>Alter the system call filter applied to containers. Takes a space-separated list of system call
 752         names or group names (the latter prefixed with <literal>@</literal>, as listed by the
 753         <command>syscall-filter</command> command of
 754         <citerefentry><refentrytitle>systemd-analyze</refentrytitle><manvolnum>1</manvolnum></citerefentry>). Passed
 755         system calls will be permitted. The list may optionally be prefixed by <literal>~</literal>, in which case all
 756         listed system calls are prohibited. If this command line option is used multiple times the configured lists are
 757         combined. If both a positive and a negative list (that is one system call list without and one with the
 758         <literal>~</literal> prefix) are configured, the negative list takes precedence over the positive list. Note
 759         that <command>systemd-nspawn</command> always implements a system call whitelist (as opposed to a blacklist),
 760         and this command line option hence adds or removes entries from the default whitelist, depending on the
 761         <literal>~</literal> prefix. Note that the applied system call filter is also altered implicitly if additional
 762         capabilities are passed using the <command>--capabilities=</command>.</para></listitem>
 763       </varlistentry>
 764
 765       <varlistentry>
 766         <term><option>--rlimit=</option></term>
 767
 768         <listitem><para>Sets the specified POSIX resource limit for the container payload. Expects an assignment of the
 769         form
 770         <literal><replaceable>LIMIT</replaceable>=<replaceable>SOFT</replaceable>:<replaceable>HARD</replaceable></literal>
 771         or <literal><replaceable>LIMIT</replaceable>=<replaceable>VALUE</replaceable></literal>, where
 772         <replaceable>LIMIT</replaceable> should refer to a resource limit type, such as
 773         <constant>RLIMIT_NOFILE</constant> or <constant>RLIMIT_NICE</constant>. The <replaceable>SOFT</replaceable> and
 774         <replaceable>HARD</replaceable> fields should refer to the numeric soft and hard resource limit values. If the
 775         second form is used, <replaceable>VALUE</replaceable> may specify a value that is used both as soft and hard
 776         limit. In place of a numeric value the special string <literal>infinity</literal> may be used to turn off
 777         resource limiting for the specific type of resource. This command line option may be used multiple times to
 778         control limits on multiple limit types. If used multiple times for the same limit type, the last use
 779         wins. For details about resource limits see <citerefentry
 780         project='man-pages'><refentrytitle>setrlimit</refentrytitle><manvolnum>2</manvolnum></citerefentry>. By default
 781         resource limits for the container's init process (PID 1) are set to the same values the Linux kernel originally
 782         passed to the host init system. Note that some resource limits are enforced on resources counted per user, in
 783         particular <constant>RLIMIT_NPROC</constant>. This means that unless user namespacing is deployed
 784         (i.e. <option>--private-users=</option> is used, see above), any limits set will be applied to the resource
 785         usage of the same user on all local containers as well as the host. This means particular care needs to be
 786         taken with these limits as they might be triggered by possibly less trusted code. Example:
 787         <literal>--rlimit=RLIMIT_NOFILE=8192:16384</literal>.</para></listitem>
 788       </varlistentry>
 789
 790       <varlistentry>
 791         <term><option>--oom-score-adjust=</option></term>
 792
 793         <listitem><para>Changes the OOM ("Out Of Memory") score adjustment value for the container payload. This controls
 794         <filename>/proc/self/oom_score_adj</filename> which influences the preference with which this container is
 795         terminated when memory becomes scarce. For details see <citerefentry
 796         project='man-pages'><refentrytitle>proc</refentrytitle><manvolnum>5</manvolnum></citerefentry>. Takes an
 797         integer in the range -1000…1000.</para></listitem>
 798       </varlistentry>
 799
 800       <varlistentry>
 801         <term><option>--cpu-affinity=</option></term>
 802
 803         <listitem><para>Controls the CPU affinity of the container payload. Takes a comma separated list of CPU numbers
 804         or number ranges (the latter's start and end value separated by dashes). See <citerefentry
 805         project='man-pages'><refentrytitle>sched_setaffinity</refentrytitle><manvolnum>2</manvolnum></citerefentry> for
 806         details.</para></listitem>
 807       </varlistentry>
 808
 809       <varlistentry>
 810         <term><option>--kill-signal=</option></term>
 811
 812         <listitem><para>Specify the process signal to send to the container's PID 1 when nspawn itself receives
 813         <constant>SIGTERM</constant>, in order to trigger an orderly shutdown of the container. Defaults to
 814         <constant>SIGRTMIN+3</constant> if <option>--boot</option> is used (on systemd-compatible init systems
 815         <constant>SIGRTMIN+3</constant> triggers an orderly shutdown). If <option>--boot</option> is not used and this
 816         option is not specified the container's processes are terminated abrubtly via <constant>SIGKILL</constant>. For
 817         a list of valid signals, see <citerefentry
 818         project='man-pages'><refentrytitle>signal</refentrytitle><manvolnum>7</manvolnum></citerefentry>.</para></listitem>
 819       </varlistentry>
 820
 821       <varlistentry>
 822         <term><option>--link-journal=</option></term>
 823
 824         <listitem><para>Control whether the container's journal shall
 825         be made visible to the host system. If enabled, allows viewing
 826         the container's journal files from the host (but not vice
 827         versa). Takes one of <literal>no</literal>,
 828         <literal>host</literal>, <literal>try-host</literal>,
 829         <literal>guest</literal>, <literal>try-guest</literal>,
 830         <literal>auto</literal>. If <literal>no</literal>, the journal
 831         is not linked. If <literal>host</literal>, the journal files
 832         are stored on the host file system (beneath
 833         <filename>/var/log/journal/<replaceable>machine-id</replaceable></filename>)
 834         and the subdirectory is bind-mounted into the container at the
 835         same location. If <literal>guest</literal>, the journal files
 836         are stored on the guest file system (beneath
 837         <filename>/var/log/journal/<replaceable>machine-id</replaceable></filename>)
 838         and the subdirectory is symlinked into the host at the same
 839         location. <literal>try-host</literal> and
 840         <literal>try-guest</literal> do the same but do not fail if
 841         the host does not have persistent journaling enabled. If
 842         <literal>auto</literal> (the default), and the right
 843         subdirectory of <filename>/var/log/journal</filename> exists,
 844         it will be bind mounted into the container. If the
 845         subdirectory does not exist, no linking is performed.
 846         Effectively, booting a container once with
 847         <literal>guest</literal> or <literal>host</literal> will link
 848         the journal persistently if further on the default of
 849         <literal>auto</literal> is used.</para>
 850
 851         <para>Note that <option>--link-journal=try-guest</option> is the default if the
 852         <filename>systemd-nspawn@.service</filename> template unit file is used.</para></listitem>
 853       </varlistentry>
 854
 855       <varlistentry>
 856         <term><option>-j</option></term>
 857
 858         <listitem><para>Equivalent to
 859         <option>--link-journal=try-guest</option>.</para></listitem>
 860       </varlistentry>
 861
 862       <varlistentry>
 863         <term><option>--resolv-conf=</option></term>
 864
 865         <listitem><para>Configures how <filename>/etc/resolv.conf</filename> inside of the container (i.e. DNS
 866         configuration synchronization from host to container) shall be handled. Takes one of <literal>off</literal>,
 867         <literal>copy-host</literal>, <literal>copy-static</literal>, <literal>bind-host</literal>,
 868         <literal>bind-static</literal>, <literal>delete</literal> or <literal>auto</literal>. If set to
 869         <literal>off</literal> the <filename>/etc/resolv.conf</filename> file in the container is left as it is
 870         included in the image, and neither modified nor bind mounted over. If set to <literal>copy-host</literal>, the
 871         <filename>/etc/resolv.conf</filename> file from the host is copied into the container. Similar, if
 872         <literal>bind-host</literal> is used, the file is bind mounted from the host into the container. If set to
 873         <literal>copy-static</literal> the static <filename>resolv.conf</filename> file supplied with
 874         <citerefentry><refentrytitle>systemd-resolved.service</refentrytitle><manvolnum>8</manvolnum></citerefentry> is
 875         copied into the container, and correspondingly <literal>bind-static</literal> bind mounts it there. If set to
 876         <literal>delete</literal> the <filename>/etc/resolv.conf</filename> file in the container is deleted if it
 877         exists. Finally, if set to <literal>auto</literal> the file is left as it is if private networking is turned on
 878         (see <option>--private-network</option>). Otherwise, if <filename>systemd-resolved.service</filename> is
 879         connectible its static <filename>resolv.conf</filename> file is used, and if not the host's
 880         <filename>/etc/resolv.conf</filename> file is used. In the latter cases the file is copied if the image is
 881         writable, and bind mounted otherwise. It's recommended to use <literal>copy</literal> if the container shall be
 882         able to make changes to the DNS configuration on its own, deviating from the host's settings. Otherwise
 883         <literal>bind</literal> is preferable, as it means direct changes to <filename>/etc/resolv.conf</filename> in
 884         the container are not allowed, as it is a read-only bind mount (but note that if the container has enough
 885         privileges, it might simply go ahead and unmount the bind mount anyway). Note that both if the file is bind
 886         mounted and if it is copied no further propagation of configuration is generally done after the one-time early
 887         initialization (this is because the file is usually updated through copying and renaming). Defaults to
 888         <literal>auto</literal>.</para></listitem>
 889       </varlistentry>
 890
 891       <varlistentry>
 892         <term><option>--timezone=</option></term>
 893
 894         <listitem><para>Configures how <filename>/etc/localtime</filename> inside of the container (i.e. local timezone
 895         synchronization from host to container) shall be handled. Takes one of <literal>off</literal>,
 896         <literal>copy</literal>, <literal>bind</literal>, <literal>symlink</literal>, <literal>delete</literal> or
 897         <literal>auto</literal>. If set to <literal>off</literal> the <filename>/etc/localtime</filename> file in the
 898         container is left as it is included in the image, and neither modified nor bind mounted over. If set to
 899         <literal>copy</literal> the <filename>/etc/localtime</filename> file of the host is copied into the
 900         container. Similar, if <literal>bind</literal> is used, it is bind mounted from the host into the container. If
 901         set to <literal>symlink</literal> a symlink from <filename>/etc/localtime</filename> in the container is
 902         created pointing to the matching the timezone file of the container that matches the timezone setting on the
 903         host. If set to <literal>delete</literal> the file in the container is deleted, should it exist. If set to
 904         <literal>auto</literal> and the <filename>/etc/localtime</filename> file of the host is a symlink, then
 905         <literal>symlink</literal> mode is used, and <literal>copy</literal> otherwise, except if the image is
 906         read-only in which case <literal>bind</literal> is used instead. Defaults to
 907         <literal>auto</literal>.</para></listitem>
 908       </varlistentry>
 909
 910       <varlistentry>
 911         <term><option>--read-only</option></term>
 912
 913         <listitem><para>Mount the root file system read-only for the
 914         container.</para></listitem>
 915       </varlistentry>
 916
 917       <varlistentry>
 918         <term><option>--bind=</option></term>
 919         <term><option>--bind-ro=</option></term>
 920
 921         <listitem><para>Bind mount a file or directory from the host into the container. Takes one of: a path
 922         argument — in which case the specified path will be mounted from the host to the same path in the container, or
 923         a colon-separated pair of paths — in which case the first specified path is the source in the host, and the
 924         second path is the destination in the container, or a colon-separated triple of source path, destination path
 925         and mount options. The source path may optionally be prefixed with a <literal>+</literal> character. If so, the
 926         source path is taken relative to the image's root directory. This permits setting up bind mounts within the
 927         container image. The source path may be specified as empty string, in which case a temporary directory below
 928         the host's <filename>/var/tmp</filename> directory is used. It is automatically removed when the container is
 929         shut down. Mount options are comma-separated and currently, only <option>rbind</option> and
 930         <option>norbind</option> are allowed, controlling whether to create a recursive or a regular bind
 931         mount. Defaults to "rbind". Backslash escapes are interpreted, so <literal>\:</literal> may be used to embed
 932         colons in either path.  This option may be specified multiple times for creating multiple independent bind
 933         mount points. The <option>--bind-ro=</option> option creates read-only bind mounts.</para>
 934
 935         <para>Note that when this option is used in combination with <option>--private-users</option>, the resulting
 936         mount points will be owned by the <constant>nobody</constant> user. That's because the mount and its files and
 937         directories continue to be owned by the relevant host users and groups, which do not exist in the container,
 938         and thus show up under the wildcard UID 65534 (nobody). If such bind mounts are created, it is recommended to
 939         make them read-only, using <option>--bind-ro=</option>.</para></listitem>
 940       </varlistentry>
 941
 942       <varlistentry>
 943         <term><option>--tmpfs=</option></term>
 944
 945         <listitem><para>Mount a tmpfs file system into the container.
 946         Takes a single absolute path argument that specifies where to
 947         mount the tmpfs instance to (in which case the directory
 948         access mode will be chosen as 0755, owned by root/root), or
 949         optionally a colon-separated pair of path and mount option
 950         string that is used for mounting (in which case the kernel
 951         default for access mode and owner will be chosen, unless
 952         otherwise specified). This option is particularly useful for
 953         mounting directories such as <filename>/var</filename> as
 954         tmpfs, to allow state-less systems, in particular when
 955         combined with <option>--read-only</option>.
 956         Backslash escapes are interpreted in the path, so
 957         <literal>\:</literal> may be used to embed colons in the path.
 958         </para></listitem>
 959       </varlistentry>
 960
 961       <varlistentry>
 962         <term><option>--overlay=</option></term>
 963         <term><option>--overlay-ro=</option></term>
 964
 965         <listitem><para>Combine multiple directory trees into one
 966         overlay file system and mount it into the container. Takes a
 967         list of colon-separated paths to the directory trees to
 968         combine and the destination mount point.</para>
 969
 970         <para>Backslash escapes are interpreted in the paths, so
 971         <literal>\:</literal> may be used to embed colons in the paths.
 972         </para>
 973
 974         <para>If three or more paths are specified, then the last
 975         specified path is the destination mount point in the
 976         container, all paths specified before refer to directory trees
 977         on the host and are combined in the specified order into one
 978         overlay file system. The left-most path is hence the lowest
 979         directory tree, the second-to-last path the highest directory
 980         tree in the stacking order. If <option>--overlay-ro=</option>
 981         is used instead of <option>--overlay=</option>, a read-only
 982         overlay file system is created. If a writable overlay file
 983         system is created, all changes made to it are written to the
 984         highest directory tree in the stacking order, i.e. the
 985         second-to-last specified.</para>
 986
 987         <para>If only two paths are specified, then the second
 988         specified path is used both as the top-level directory tree in
 989         the stacking order as seen from the host, as well as the mount
 990         point for the overlay file system in the container. At least
 991         two paths have to be specified.</para>
 992
 993         <para>The source paths may optionally be prefixed with <literal>+</literal> character. If so they are taken
 994         relative to the image's root directory. The uppermost source path may also be specified as empty string, in
 995         which case a temporary directory below the host's <filename>/var/tmp</filename> is used. The directory is
 996         removed automatically when the container is shut down. This behaviour is useful in order to make read-only
 997         container directories writable while the container is running. For example, use the
 998         <literal>--overlay=+/var::/var</literal> option in order to automatically overlay a writable temporary
 999         directory on a read-only <filename>/var</filename> directory.</para>
1000
1001         <para>For details about overlay file systems, see <ulink
1002         url="https://www.kernel.org/doc/Documentation/filesystems/overlayfs.txt">overlayfs.txt</ulink>. Note
1003         that the semantics of overlay file systems are substantially
1004         different from normal file systems, in particular regarding
1005         reported device and inode information. Device and inode
1006         information may change for a file while it is being written
1007         to, and processes might see out-of-date versions of files at
1008         times. Note that this switch automatically derives the
1009         <literal>workdir=</literal> mount option for the overlay file
1010         system from the top-level directory tree, making it a sibling
1011         of it. It is hence essential that the top-level directory tree
1012         is not a mount point itself (since the working directory must
1013         be on the same file system as the top-most directory
1014         tree). Also note that the <literal>lowerdir=</literal> mount
1015         option receives the paths to stack in the opposite order of
1016         this switch.</para></listitem>
1017       </varlistentry>
1018
1019       <varlistentry>
1020         <term><option>-E <replaceable>NAME</replaceable>=<replaceable>VALUE</replaceable></option></term>
1021         <term><option>--setenv=<replaceable>NAME</replaceable>=<replaceable>VALUE</replaceable></option></term>
1022
1023         <listitem><para>Specifies an environment variable assignment
1024         to pass to the init process in the container, in the format
1025         <literal>NAME=VALUE</literal>. This may be used to override
1026         the default variables or to set additional variables. This
1027         parameter may be used more than once.</para></listitem>
1028       </varlistentry>
1029
1030       <varlistentry>
1031         <term><option>--register=</option></term>
1032
1033         <listitem><para>Controls whether the container is registered with
1034         <citerefentry><refentrytitle>systemd-machined</refentrytitle><manvolnum>8</manvolnum></citerefentry>.  Takes a
1035         boolean argument, which defaults to <literal>yes</literal>.  This option should be enabled when the container
1036         runs a full Operating System (more specifically: a system and service manager as PID 1), and is useful to
1037         ensure that the container is accessible via
1038         <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry> and shown by
1039         tools such as <citerefentry
1040         project='man-pages'><refentrytitle>ps</refentrytitle><manvolnum>1</manvolnum></citerefentry>.  If the container
1041         does not run a service manager, it is recommended to set this option to
1042         <literal>no</literal>.</para></listitem>
1043       </varlistentry>
1044
1045       <varlistentry>
1046         <term><option>--keep-unit</option></term>
1047
1048         <listitem><para>Instead of creating a transient scope unit to run the container in, simply use the service or
1049         scope unit <command>systemd-nspawn</command> has been invoked in. If <option>--register=yes</option> is set
1050         this unit is registered with
1051         <citerefentry><refentrytitle>systemd-machined</refentrytitle><manvolnum>8</manvolnum></citerefentry>. This
1052         switch should be used if <command>systemd-nspawn</command> is invoked from within a service unit, and the
1053         service unit's sole purpose is to run a single <command>systemd-nspawn</command> container. This option is not
1054         available if run from a user session.</para>
1055         <para>Note that passing <option>--keep-unit</option> disables the effect of <option>--slice=</option> and
1056         <option>--property=</option>. Use <option>--keep-unit</option> and <option>--register=no</option> in
1057         combination to disable any kind of unit allocation or registration with
1058         <command>systemd-machined</command>.</para></listitem>
1059       </varlistentry>
1060
1061       <varlistentry>
1062         <term><option>--personality=</option></term>
1063
1064         <listitem><para>Control the architecture ("personality")
1065         reported by
1066         <citerefentry project='man-pages'><refentrytitle>uname</refentrytitle><manvolnum>2</manvolnum></citerefentry>
1067         in the container. Currently, only <literal>x86</literal> and
1068         <literal>x86-64</literal> are supported. This is useful when
1069         running a 32-bit container on a 64-bit host. If this setting
1070         is not used, the personality reported in the container is the
1071         same as the one reported on the host.</para></listitem>
1072       </varlistentry>
1073
1074       <varlistentry>
1075         <term><option>-q</option></term>
1076         <term><option>--quiet</option></term>
1077
1078         <listitem><para>Turns off any status output by the tool
1079         itself. When this switch is used, the only output from nspawn
1080         will be the console output of the container OS
1081         itself.</para></listitem>
1082       </varlistentry>
1083
1084       <varlistentry>
1085         <term><option>--volatile</option></term>
1086         <term><option>--volatile=</option><replaceable>MODE</replaceable></term>
1087
1088         <listitem><para>Boots the container in volatile mode. When no
1089         mode parameter is passed or when mode is specified as
1090         <option>yes</option>, full volatile mode is enabled. This
1091         means the root directory is mounted as a mostly unpopulated
1092         <literal>tmpfs</literal> instance, and
1093         <filename>/usr</filename> from the OS tree is mounted into it
1094         in read-only mode (the system thus starts up with read-only OS
1095         image, but pristine state and configuration, any changes
1096         are lost on shutdown). When the mode parameter
1097         is specified as <option>state</option>, the OS tree is
1098         mounted read-only, but <filename>/var</filename> is mounted as
1099         a <literal>tmpfs</literal> instance into it (the system thus
1100         starts up with read-only OS resources and configuration, but
1101         pristine state, and any changes to the latter are lost on
1102         shutdown). When the mode parameter is specified as
1103         <option>no</option> (the default), the whole OS tree is made
1104         available writable.</para>
1105
1106         <para>This option provides similar functionality for containers as the <literal>systemd.volatile=</literal>
1107         kernel command line switch provides for host systems. See
1108         <citerefentry><refentrytitle>kernel-command-line</refentrytitle><manvolnum>7</manvolnum></citerefentry> for
1109         details.</para>
1110
1111         <para>Note that enabling this setting will only work correctly with operating systems in the container that can
1112         boot up with only <filename>/usr</filename> mounted, and are able to automatically populate
1113         <filename>/var</filename>, and also <filename>/etc</filename> in case of
1114         <literal>--volatile=yes</literal>.</para></listitem>
1115       </varlistentry>
1116
1117       <varlistentry>
1118         <term><option>--settings=</option><replaceable>MODE</replaceable></term>
1119
1120         <listitem><para>Controls whether
1121         <command>systemd-nspawn</command> shall search for and use
1122         additional per-container settings from
1123         <filename>.nspawn</filename> files. Takes a boolean or the
1124         special values <option>override</option> or
1125         <option>trusted</option>.</para>
1126
1127         <para>If enabled (the default), a settings file named after the
1128         machine (as specified with the <option>--machine=</option>
1129         setting, or derived from the directory or image file name)
1130         with the suffix <filename>.nspawn</filename> is searched in
1131         <filename>/etc/systemd/nspawn/</filename> and
1132         <filename>/run/systemd/nspawn/</filename>. If it is found
1133         there, its settings are read and used. If it is not found
1134         there, it is subsequently searched in the same directory as the
1135         image file or in the immediate parent of the root directory of
1136         the container. In this case, if the file is found, its settings
1137         will be also read and used, but potentially unsafe settings
1138         are ignored. Note that in both these cases, settings on the
1139         command line take precedence over the corresponding settings
1140         from loaded <filename>.nspawn</filename> files, if both are
1141         specified. Unsafe settings are considered all settings that
1142         elevate the container's privileges or grant access to
1143         additional resources such as files or directories of the
1144         host. For details about the format and contents of
1145         <filename>.nspawn</filename> files, consult
1146         <citerefentry><refentrytitle>systemd.nspawn</refentrytitle><manvolnum>5</manvolnum></citerefentry>.</para>
1147
1148         <para>If this option is set to <option>override</option>, the
1149         file is searched, read and used the same way, however, the order of
1150         precedence is reversed: settings read from the
1151         <filename>.nspawn</filename> file will take precedence over
1152         the corresponding command line options, if both are
1153         specified.</para>
1154
1155         <para>If this option is set to <option>trusted</option>, the
1156         file is searched, read and used the same way, but regardless
1157         of being found in <filename>/etc/systemd/nspawn/</filename>,
1158         <filename>/run/systemd/nspawn/</filename> or next to the image
1159         file or container root directory, all settings will take
1160         effect, however, command line arguments still take precedence
1161         over corresponding settings.</para>
1162
1163         <para>If disabled, no <filename>.nspawn</filename> file is read
1164         and no settings except the ones on the command line are in
1165         effect.</para></listitem>
1166       </varlistentry>
1167
1168       <varlistentry>
1169         <term><option>--notify-ready=</option></term>
1170
1171         <listitem><para>Configures support for notifications from the container's init process.
1172         <option>--notify-ready=</option> takes a boolean (<option>no</option> and  <option>yes</option>).
1173         With option <option>no</option> systemd-nspawn notifies systemd
1174         with a <literal>READY=1</literal> message when the init process is created.
1175         With option <option>yes</option> systemd-nspawn waits for the
1176         <literal>READY=1</literal> message from the init process in the container
1177         before sending its own to systemd. For more details about notifications
1178         see <citerefentry><refentrytitle>sd_notify</refentrytitle><manvolnum>3</manvolnum></citerefentry>).</para></listitem>
1179       </varlistentry>
1180
1181       <xi:include href="standard-options.xml" xpointer="help" />
1182       <xi:include href="standard-options.xml" xpointer="version" />
1183     </variablelist>
1184
1185   </refsect1>
1186
1187   <refsect1>
1188     <title>Examples</title>
1189
1190     <example>
1191       <title>Download a
1192       <ulink url="https://getfedora.org">Fedora</ulink> image and start a shell in it</title>
1193
1194       <programlisting># machinectl pull-raw --verify=no \
1195       https://download.fedoraproject.org/pub/fedora/linux/releases/&fedora_latest_version;/Cloud/x86_64/images/Fedora-Cloud-Base-&fedora_latest_version;-&fedora_cloud_release;.x86_64.raw.xz
1196 # systemd-nspawn -M Fedora-Cloud-Base-&fedora_latest_version;-&fedora_cloud_release;.x86_64.raw</programlisting>
1197
1198       <para>This downloads an image using
1199       <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry>
1200       and opens a shell in it.</para>
1201     </example>
1202
1203     <example>
1204       <title>Build and boot a minimal Fedora distribution in a container</title>
1205
1206       <programlisting># dnf -y --releasever=&fedora_latest_version; --installroot=/var/lib/machines/f&fedora_latest_version; \
1207       --disablerepo='*' --enablerepo=fedora --enablerepo=updates install \
1208       systemd passwd dnf fedora-release vim-minimal
1209 # systemd-nspawn -bD /var/lib/machines/f&fedora_latest_version;</programlisting>
1210
1211       <para>This installs a minimal Fedora distribution into the
1212       directory <filename noindex='true'>/var/lib/machines/f&fedora_latest_version;</filename>
1213       and then boots an OS in a namespace container in it. Because the installation
1214       is located underneath the standard <filename>/var/lib/machines/</filename>
1215       directory, it is also possible to start the machine using
1216       <command>systemd-nspawn -M f&fedora_latest_version;</command>.</para>
1217     </example>
1218
1219     <example>
1220       <title>Spawn a shell in a container of a minimal Debian unstable distribution</title>
1221
1222       <programlisting># debootstrap unstable ~/debian-tree/
1223 # systemd-nspawn -D ~/debian-tree/</programlisting>
1224
1225       <para>This installs a minimal Debian unstable distribution into
1226       the directory <filename>~/debian-tree/</filename> and then
1227       spawns a shell in a namespace container in it.</para>
1228
1229       <para><command>debootstrap</command> supports
1230       <ulink url="https://www.debian.org">Debian</ulink>,
1231       <ulink url="https://www.ubuntu.com">Ubuntu</ulink>,
1232       and <ulink url="https://www.tanglu.org">Tanglu</ulink>
1233       out of the box, so the same command can be used to install any of those. For other
1234       distributions from the Debian family, a mirror has to be specified, see
1235       <citerefentry project='die-net'><refentrytitle>debootstrap</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
1236       </para>
1237     </example>
1238
1239     <example>
1240       <title>Boot a minimal
1241       <ulink url="https://www.archlinux.org">Arch Linux</ulink> distribution in a container</title>
1242
1243       <programlisting># pacstrap -c -d ~/arch-tree/ base
1244 # systemd-nspawn -bD ~/arch-tree/</programlisting>
1245
1246       <para>This installs a minimal Arch Linux distribution into the
1247       directory <filename>~/arch-tree/</filename> and then boots an OS
1248       in a namespace container in it.</para>
1249     </example>
1250
1251     <example>
1252       <title>Install the
1253       <ulink url="https://software.opensuse.org/distributions/tumbleweed">OpenSUSE Tumbleweed</ulink>
1254       rolling distribution</title>
1255
1256       <programlisting># zypper --root=/var/lib/machines/tumbleweed ar -c \
1257       https://download.opensuse.org/tumbleweed/repo/oss tumbleweed
1258 # zypper --root=/var/lib/machines/tumbleweed refresh
1259 # zypper --root=/var/lib/machines/tumbleweed install --no-recommends \
1260       systemd shadow zypper openSUSE-release vim
1261 # systemd-nspawn -M tumbleweed passwd root
1262 # systemd-nspawn -M tumbleweed -b</programlisting>
1263     </example>
1264
1265     <example>
1266       <title>Boot into an ephemeral snapshot of the host system</title>
1267
1268       <programlisting># systemd-nspawn -D / -xb</programlisting>
1269
1270       <para>This runs a copy of the host system in a snapshot which is removed immediately when the container
1271       exits. All file system changes made during runtime will be lost on shutdown, hence.</para>
1272     </example>
1273
1274     <example>
1275       <title>Run a container with SELinux sandbox security contexts</title>
1276
1277       <programlisting># chcon system_u:object_r:svirt_sandbox_file_t:s0:c0,c1 -R /srv/container
1278 # systemd-nspawn -L system_u:object_r:svirt_sandbox_file_t:s0:c0,c1 \
1279       -Z system_u:system_r:svirt_lxc_net_t:s0:c0,c1 -D /srv/container /bin/sh</programlisting>
1280     </example>
1281
1282     <example>
1283       <title>Run a container with an OSTree deployment</title>
1284
1285       <programlisting># systemd-nspawn -b -i ~/image.raw \
1286       --pivot-root=/ostree/deploy/$OS/deploy/$CHECKSUM:/sysroot \
1287       --bind=+/sysroot/ostree/deploy/$OS/var:/var</programlisting>
1288     </example>
1289   </refsect1>
1290
1291   <refsect1>
1292     <title>Exit status</title>
1293
1294     <para>The exit code of the program executed in the container is
1295     returned.</para>
1296   </refsect1>
1297
1298   <refsect1>
1299     <title>See Also</title>
1300     <para>
1301       <citerefentry><refentrytitle>systemd</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
1302       <citerefentry><refentrytitle>systemd.nspawn</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
1303       <citerefentry project='man-pages'><refentrytitle>chroot</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
1304       <citerefentry project='mankier'><refentrytitle>dnf</refentrytitle><manvolnum>8</manvolnum></citerefentry>,
1305       <citerefentry project='die-net'><refentrytitle>debootstrap</refentrytitle><manvolnum>8</manvolnum></citerefentry>,
1306       <citerefentry project='archlinux'><refentrytitle>pacman</refentrytitle><manvolnum>8</manvolnum></citerefentry>,
1307       <citerefentry project='mankier'><refentrytitle>zypper</refentrytitle><manvolnum>8</manvolnum></citerefentry>,
1308       <citerefentry><refentrytitle>systemd.slice</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
1309       <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
1310       <citerefentry project='man-pages'><refentrytitle>btrfs</refentrytitle><manvolnum>8</manvolnum></citerefentry>
1311     </para>
1312   </refsect1>
1313
1314 </refentry>