]>
Commit | Line | Data |
---|---|---|
f757855e | 1 | <?xml version='1.0'?> <!--*- Mode: nxml; nxml-child-indent: 2; indent-tabs-mode: nil -*--> |
8f7a3c14 | 2 | <!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN" |
12b42c76 | 3 | "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd"> |
8f7a3c14 LP |
4 | |
5 | <!-- | |
572eb058 ZJS |
6 | SPDX-License-Identifier: LGPL-2.1+ |
7 | ||
8f7a3c14 LP |
8 | This file is part of systemd. |
9 | ||
10 | Copyright 2010 Lennart Poettering | |
11 | ||
12 | systemd is free software; you can redistribute it and/or modify it | |
5430f7f2 LP |
13 | under the terms of the GNU Lesser General Public License as published by |
14 | the Free Software Foundation; either version 2.1 of the License, or | |
8f7a3c14 LP |
15 | (at your option) any later version. |
16 | ||
17 | systemd is distributed in the hope that it will be useful, but | |
18 | WITHOUT ANY WARRANTY; without even the implied warranty of | |
19 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | |
5430f7f2 | 20 | Lesser General Public License for more details. |
8f7a3c14 | 21 | |
5430f7f2 | 22 | You should have received a copy of the GNU Lesser General Public License |
8f7a3c14 LP |
23 | along with systemd; If not, see <http://www.gnu.org/licenses/>. |
24 | --> | |
25 | ||
dfdebb1b | 26 | <refentry id="systemd-nspawn" |
798d3a52 ZJS |
27 | xmlns:xi="http://www.w3.org/2001/XInclude"> |
28 | ||
29 | <refentryinfo> | |
30 | <title>systemd-nspawn</title> | |
31 | <productname>systemd</productname> | |
32 | ||
33 | <authorgroup> | |
34 | <author> | |
35 | <contrib>Developer</contrib> | |
36 | <firstname>Lennart</firstname> | |
37 | <surname>Poettering</surname> | |
38 | <email>lennart@poettering.net</email> | |
39 | </author> | |
40 | </authorgroup> | |
41 | </refentryinfo> | |
42 | ||
43 | <refmeta> | |
44 | <refentrytitle>systemd-nspawn</refentrytitle> | |
45 | <manvolnum>1</manvolnum> | |
46 | </refmeta> | |
47 | ||
48 | <refnamediv> | |
49 | <refname>systemd-nspawn</refname> | |
50 | <refpurpose>Spawn a namespace container for debugging, testing and building</refpurpose> | |
51 | </refnamediv> | |
52 | ||
53 | <refsynopsisdiv> | |
54 | <cmdsynopsis> | |
55 | <command>systemd-nspawn</command> | |
56 | <arg choice="opt" rep="repeat">OPTIONS</arg> | |
57 | <arg choice="opt"><replaceable>COMMAND</replaceable> | |
58 | <arg choice="opt" rep="repeat">ARGS</arg> | |
59 | </arg> | |
60 | </cmdsynopsis> | |
61 | <cmdsynopsis> | |
62 | <command>systemd-nspawn</command> | |
4447e799 | 63 | <arg choice="plain">--boot</arg> |
798d3a52 ZJS |
64 | <arg choice="opt" rep="repeat">OPTIONS</arg> |
65 | <arg choice="opt" rep="repeat">ARGS</arg> | |
66 | </cmdsynopsis> | |
67 | </refsynopsisdiv> | |
68 | ||
69 | <refsect1> | |
70 | <title>Description</title> | |
71 | ||
b09c0bba LP |
72 | <para><command>systemd-nspawn</command> may be used to run a command or OS in a light-weight namespace |
73 | container. In many ways it is similar to <citerefentry | |
74 | project='man-pages'><refentrytitle>chroot</refentrytitle><manvolnum>1</manvolnum></citerefentry>, but more powerful | |
75 | since it fully virtualizes the file system hierarchy, as well as the process tree, the various IPC subsystems and | |
76 | the host and domain name.</para> | |
77 | ||
5164c3b4 | 78 | <para><command>systemd-nspawn</command> may be invoked on any directory tree containing an operating system tree, |
b09c0bba | 79 | using the <option>--directory=</option> command line option. By using the <option>--machine=</option> option an OS |
5164c3b4 | 80 | tree is automatically searched for in a couple of locations, most importantly in |
b09c0bba LP |
81 | <filename>/var/lib/machines</filename>, the suggested directory to place container images installed on the |
82 | system.</para> | |
83 | ||
84 | <para>In contrast to <citerefentry | |
85 | project='man-pages'><refentrytitle>chroot</refentrytitle><manvolnum>1</manvolnum></citerefentry> <command>systemd-nspawn</command> | |
86 | may be used to boot full Linux-based operating systems in a container.</para> | |
87 | ||
88 | <para><command>systemd-nspawn</command> limits access to various kernel interfaces in the container to read-only, | |
89 | such as <filename>/sys</filename>, <filename>/proc/sys</filename> or <filename>/sys/fs/selinux</filename>. The | |
90 | host's network interfaces and the system clock may not be changed from within the container. Device nodes may not | |
91 | be created. The host system cannot be rebooted and kernel modules may not be loaded from within the | |
798d3a52 ZJS |
92 | container.</para> |
93 | ||
b09c0bba LP |
94 | <para>Use a tool like <citerefentry |
95 | project='mankier'><refentrytitle>dnf</refentrytitle><manvolnum>8</manvolnum></citerefentry>, <citerefentry | |
96 | project='die-net'><refentrytitle>debootstrap</refentrytitle><manvolnum>8</manvolnum></citerefentry>, or | |
97 | <citerefentry project='archlinux'><refentrytitle>pacman</refentrytitle><manvolnum>8</manvolnum></citerefentry> to | |
98 | set up an OS directory tree suitable as file system hierarchy for <command>systemd-nspawn</command> containers. See | |
99 | the Examples section below for details on suitable invocation of these commands.</para> | |
100 | ||
101 | <para>As a safety check <command>systemd-nspawn</command> will verify the existence of | |
102 | <filename>/usr/lib/os-release</filename> or <filename>/etc/os-release</filename> in the container tree before | |
103 | starting the container (see | |
104 | <citerefentry><refentrytitle>os-release</refentrytitle><manvolnum>5</manvolnum></citerefentry>). It might be | |
105 | necessary to add this file to the container tree manually if the OS of the container is too old to contain this | |
798d3a52 | 106 | file out-of-the-box.</para> |
b09c0bba LP |
107 | |
108 | <para><command>systemd-nspawn</command> may be invoked directly from the interactive command line or run as system | |
109 | service in the background. In this mode each container instance runs as its own service instance; a default | |
110 | template unit file <filename>systemd-nspawn@.service</filename> is provided to make this easy, taking the container | |
111 | name as instance identifier. Note that different default options apply when <command>systemd-nspawn</command> is | |
6dd6a9c4 | 112 | invoked by the template unit file than interactively on the command line. Most importantly the template unit file |
b09c0bba | 113 | makes use of the <option>--boot</option> which is not the default in case <command>systemd-nspawn</command> is |
6dd6a9c4 | 114 | invoked from the interactive command line. Further differences with the defaults are documented along with the |
b09c0bba LP |
115 | various supported options below.</para> |
116 | ||
117 | <para>The <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry> tool may | |
118 | be used to execute a number of operations on containers. In particular it provides easy-to-use commands to run | |
119 | containers as system services using the <filename>systemd-nspawn@.service</filename> template unit | |
120 | file.</para> | |
121 | ||
122 | <para>Along with each container a settings file with the <filename>.nspawn</filename> suffix may exist, containing | |
123 | additional settings to apply when running the container. See | |
124 | <citerefentry><refentrytitle>systemd.nspawn</refentrytitle><manvolnum>5</manvolnum></citerefentry> for | |
125 | details. Settings files override the default options used by the <filename>systemd-nspawn@.service</filename> | |
126 | template unit file, making it usually unnecessary to alter this template file directly.</para> | |
127 | ||
128 | <para>Note that <command>systemd-nspawn</command> will mount file systems private to the container to | |
129 | <filename>/dev</filename>, <filename>/run</filename> and similar. These will not be visible outside of the | |
130 | container, and their contents will be lost when the container exits.</para> | |
131 | ||
132 | <para>Note that running two <command>systemd-nspawn</command> containers from the same directory tree will not make | |
133 | processes in them see each other. The PID namespace separation of the two containers is complete and the containers | |
134 | will share very few runtime objects except for the underlying file system. Use | |
135 | <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry>'s | |
136 | <command>login</command> or <command>shell</command> commands to request an additional login session in a running | |
137 | container.</para> | |
138 | ||
139 | <para><command>systemd-nspawn</command> implements the <ulink | |
28a0ad81 | 140 | url="https://www.freedesktop.org/wiki/Software/systemd/ContainerInterface">Container Interface</ulink> |
b09c0bba LP |
141 | specification.</para> |
142 | ||
143 | <para>While running, containers invoked with <command>systemd-nspawn</command> are registered with the | |
144 | <citerefentry><refentrytitle>systemd-machined</refentrytitle><manvolnum>8</manvolnum></citerefentry> service that | |
145 | keeps track of running containers, and provides programming interfaces to interact with them.</para> | |
798d3a52 ZJS |
146 | </refsect1> |
147 | ||
148 | <refsect1> | |
149 | <title>Options</title> | |
150 | ||
151 | <para>If option <option>-b</option> is specified, the arguments | |
152 | are used as arguments for the init binary. Otherwise, | |
153 | <replaceable>COMMAND</replaceable> specifies the program to launch | |
154 | in the container, and the remaining arguments are used as | |
b09c0bba | 155 | arguments for this program. If <option>--boot</option> is not used and |
ff9b60f3 | 156 | no arguments are specified, a shell is launched in the |
798d3a52 ZJS |
157 | container.</para> |
158 | ||
159 | <para>The following options are understood:</para> | |
160 | ||
161 | <variablelist> | |
162 | <varlistentry> | |
163 | <term><option>-D</option></term> | |
164 | <term><option>--directory=</option></term> | |
165 | ||
166 | <listitem><para>Directory to use as file system root for the | |
167 | container.</para> | |
168 | ||
169 | <para>If neither <option>--directory=</option>, nor | |
170 | <option>--image=</option> is specified the directory is | |
32b64cce RM |
171 | determined by searching for a directory named the same as the |
172 | machine name specified with <option>--machine=</option>. See | |
173 | <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry> | |
174 | section "Files and Directories" for the precise search path.</para> | |
175 | ||
176 | <para>If neither <option>--directory=</option>, | |
177 | <option>--image=</option>, nor <option>--machine=</option> | |
178 | are specified, the current directory will | |
179 | be used. May not be specified together with | |
798d3a52 ZJS |
180 | <option>--image=</option>.</para></listitem> |
181 | </varlistentry> | |
182 | ||
183 | <varlistentry> | |
184 | <term><option>--template=</option></term> | |
185 | ||
17cbb288 LP |
186 | <listitem><para>Directory or <literal>btrfs</literal> subvolume to use as template for the container's root |
187 | directory. If this is specified and the container's root directory (as configured by | |
188 | <option>--directory=</option>) does not yet exist it is created as <literal>btrfs</literal> snapshot (if | |
189 | supported) or plain directory (otherwise) and populated from this template tree. Ideally, the specified | |
190 | template path refers to the root of a <literal>btrfs</literal> subvolume, in which case a simple copy-on-write | |
191 | snapshot is taken, and populating the root directory is instant. If the specified template path does not refer | |
192 | to the root of a <literal>btrfs</literal> subvolume (or not even to a <literal>btrfs</literal> file system at | |
193 | all), the tree is copied (though possibly in a copy-on-write scheme — if the file system supports that), which | |
194 | can be substantially more time-consuming. May not be specified together with <option>--image=</option> or | |
3fe22bb4 LP |
195 | <option>--ephemeral</option>.</para> |
196 | ||
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> | |
798d3a52 ZJS |
200 | </varlistentry> |
201 | ||
202 | <varlistentry> | |
203 | <term><option>-x</option></term> | |
204 | <term><option>--ephemeral</option></term> | |
205 | ||
0f3be6ca LP |
206 | <listitem><para>If specified, the container is run with a temporary snapshot of its file system that is removed |
207 | immediately when the container terminates. May not be specified together with | |
3fe22bb4 LP |
208 | <option>--template=</option>.</para> |
209 | <para>Note that this switch leaves host name, machine ID and | |
210 | all other settings that could identify the instance | |
211 | unmodified.</para></listitem> | |
798d3a52 ZJS |
212 | </varlistentry> |
213 | ||
214 | <varlistentry> | |
215 | <term><option>-i</option></term> | |
216 | <term><option>--image=</option></term> | |
217 | ||
218 | <listitem><para>Disk image to mount the root directory for the | |
219 | container from. Takes a path to a regular file or to a block | |
220 | device node. The file or block device must contain | |
221 | either:</para> | |
222 | ||
223 | <itemizedlist> | |
224 | <listitem><para>An MBR partition table with a single | |
225 | partition of type 0x83 that is marked | |
226 | bootable.</para></listitem> | |
227 | ||
228 | <listitem><para>A GUID partition table (GPT) with a single | |
229 | partition of type | |
230 | 0fc63daf-8483-4772-8e79-3d69d8477de4.</para></listitem> | |
231 | ||
232 | <listitem><para>A GUID partition table (GPT) with a marked | |
233 | root partition which is mounted as the root directory of the | |
234 | container. Optionally, GPT images may contain a home and/or | |
235 | a server data partition which are mounted to the appropriate | |
236 | places in the container. All these partitions must be | |
237 | identified by the partition types defined by the <ulink | |
28a0ad81 | 238 | url="https://www.freedesktop.org/wiki/Specifications/DiscoverablePartitionsSpec/">Discoverable |
798d3a52 | 239 | Partitions Specification</ulink>.</para></listitem> |
58abb66f LP |
240 | |
241 | <listitem><para>No partition table, and a single file system spanning the whole image.</para></listitem> | |
798d3a52 ZJS |
242 | </itemizedlist> |
243 | ||
0f3be6ca LP |
244 | <para>On GPT images, if an EFI System Partition (ESP) is discovered, it is automatically mounted to |
245 | <filename>/efi</filename> (or <filename>/boot</filename> as fallback) in case a directory by this name exists | |
246 | and is empty.</para> | |
247 | ||
58abb66f LP |
248 | <para>Partitions encrypted with LUKS are automatically decrypted. Also, on GPT images dm-verity data integrity |
249 | hash partitions are set up if the root hash for them is specified using the <option>--root-hash=</option> | |
250 | option.</para> | |
251 | ||
0f3be6ca LP |
252 | <para>Any other partitions, such as foreign partitions or swap partitions are not mounted. May not be specified |
253 | together with <option>--directory=</option>, <option>--template=</option>.</para></listitem> | |
798d3a52 | 254 | </varlistentry> |
58abb66f LP |
255 | |
256 | <varlistentry> | |
257 | <term><option>--root-hash=</option></term> | |
258 | ||
259 | <listitem><para>Takes a data integrity (dm-verity) root hash specified in hexadecimal. This option enables data | |
260 | integrity checks using dm-verity, if the used image contains the appropriate integrity data (see above). The | |
ef3116b5 | 261 | specified hash must match the root hash of integrity data, and is usually at least 256 bits (and hence 64 |
41488e1f LP |
262 | formatted hexadecimal characters) long (in case of SHA256 for example). If this option is not specified, but |
263 | the image file carries the <literal>user.verity.roothash</literal> extended file attribute (see <citerefentry | |
264 | project='man-pages'><refentrytitle>xattr</refentrytitle><manvolnum>7</manvolnum></citerefentry>), then the root | |
265 | hash is read from it, also as formatted hexadecimal characters. If the extended file attribute is not found (or | |
ef3116b5 ZJS |
266 | is not supported by the underlying file system), but a file with the <filename>.roothash</filename> suffix is |
267 | found next to the image file, bearing otherwise the same name, the root hash is read from it and automatically | |
268 | used, also as formatted hexadecimal characters.</para></listitem> | |
58abb66f | 269 | </varlistentry> |
798d3a52 | 270 | |
7732f92b LP |
271 | <varlistentry> |
272 | <term><option>-a</option></term> | |
273 | <term><option>--as-pid2</option></term> | |
274 | ||
275 | <listitem><para>Invoke the shell or specified program as process ID (PID) 2 instead of PID 1 (init). By | |
276 | default, if neither this option nor <option>--boot</option> is used, the selected binary is run as process with | |
277 | PID 1, a mode only suitable for programs that are aware of the special semantics that the process with PID 1 | |
278 | has on UNIX. For example, it needs to reap all processes reparented to it, and should implement | |
279 | <command>sysvinit</command> compatible signal handling (specifically: it needs to reboot on SIGINT, reexecute | |
280 | on SIGTERM, reload configuration on SIGHUP, and so on). With <option>--as-pid2</option> a minimal stub init | |
281 | process is run as PID 1 and the selected binary is executed as PID 2 (and hence does not need to implement any | |
282 | special semantics). The stub init process will reap processes as necessary and react appropriately to | |
283 | signals. It is recommended to use this mode to invoke arbitrary commands in containers, unless they have been | |
284 | modified to run correctly as PID 1. Or in other words: this switch should be used for pretty much all commands, | |
285 | except when the command refers to an init or shell implementation, as these are generally capable of running | |
a6b5216c | 286 | correctly as PID 1. This option may not be combined with <option>--boot</option>.</para> |
7732f92b LP |
287 | </listitem> |
288 | </varlistentry> | |
289 | ||
798d3a52 ZJS |
290 | <varlistentry> |
291 | <term><option>-b</option></term> | |
292 | <term><option>--boot</option></term> | |
293 | ||
7732f92b LP |
294 | <listitem><para>Automatically search for an init binary and invoke it as PID 1, instead of a shell or a user |
295 | supplied program. If this option is used, arguments specified on the command line are used as arguments for the | |
a6b5216c | 296 | init binary. This option may not be combined with <option>--as-pid2</option>.</para> |
7732f92b LP |
297 | |
298 | <para>The following table explains the different modes of invocation and relationship to | |
299 | <option>--as-pid2</option> (see above):</para> | |
300 | ||
301 | <table> | |
302 | <title>Invocation Mode</title> | |
303 | <tgroup cols='2' align='left' colsep='1' rowsep='1'> | |
304 | <colspec colname="switch" /> | |
305 | <colspec colname="explanation" /> | |
306 | <thead> | |
307 | <row> | |
308 | <entry>Switch</entry> | |
309 | <entry>Explanation</entry> | |
310 | </row> | |
311 | </thead> | |
312 | <tbody> | |
313 | <row> | |
314 | <entry>Neither <option>--as-pid2</option> nor <option>--boot</option> specified</entry> | |
4447e799 | 315 | <entry>The passed parameters are interpreted as the command line, which is executed as PID 1 in the container.</entry> |
7732f92b LP |
316 | </row> |
317 | ||
318 | <row> | |
319 | <entry><option>--as-pid2</option> specified</entry> | |
4447e799 | 320 | <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> |
7732f92b LP |
321 | </row> |
322 | ||
323 | <row> | |
324 | <entry><option>--boot</option> specified</entry> | |
325 | <entry>An init binary as automatically searched and run as PID 1 in the container. The passed parameters are used as invocation parameters for this process.</entry> | |
326 | </row> | |
327 | ||
328 | </tbody> | |
329 | </tgroup> | |
330 | </table> | |
b09c0bba LP |
331 | |
332 | <para>Note that <option>--boot</option> is the default mode of operation if the | |
333 | <filename>systemd-nspawn@.service</filename> template unit file is used.</para> | |
7732f92b | 334 | </listitem> |
798d3a52 ZJS |
335 | </varlistentry> |
336 | ||
5f932eb9 LP |
337 | <varlistentry> |
338 | <term><option>--chdir=</option></term> | |
339 | ||
340 | <listitem><para>Change to the specified working directory before invoking the process in the container. Expects | |
341 | an absolute path in the container's file system namespace.</para></listitem> | |
342 | </varlistentry> | |
343 | ||
b53ede69 PW |
344 | <varlistentry> |
345 | <term><option>--pivot-root=</option></term> | |
346 | ||
347 | <listitem><para>Pivot the specified directory to <filename>/</filename> inside the container, and either unmount the | |
348 | container's old root, or pivot it to another specified directory. Takes one of: a path argument — in which case the | |
349 | specified path will be pivoted to <filename>/</filename> and the old root will be unmounted; or a colon-separated pair | |
350 | of new root path and pivot destination for the old root. The new root path will be pivoted to <filename>/</filename>, | |
351 | and the old <filename>/</filename> will be pivoted to the other directory. Both paths must be absolute, and are resolved | |
352 | in the container's file system namespace.</para> | |
353 | ||
354 | <para>This is for containers which have several bootable directories in them; for example, several | |
3797fd0a ZJS |
355 | <ulink url="https://ostree.readthedocs.io/en/latest/">OSTree</ulink> deployments. It emulates the behavior of |
356 | the boot loader and initial RAM disk which normally select which directory to mount as the root and start the | |
357 | container's PID 1 in.</para></listitem> | |
b53ede69 PW |
358 | </varlistentry> |
359 | ||
798d3a52 ZJS |
360 | <varlistentry> |
361 | <term><option>-u</option></term> | |
362 | <term><option>--user=</option></term> | |
363 | ||
364 | <listitem><para>After transitioning into the container, change | |
365 | to the specified user-defined in the container's user | |
366 | database. Like all other systemd-nspawn features, this is not | |
367 | a security feature and provides protection against accidental | |
368 | destructive operations only.</para></listitem> | |
369 | </varlistentry> | |
370 | ||
371 | <varlistentry> | |
372 | <term><option>-M</option></term> | |
373 | <term><option>--machine=</option></term> | |
374 | ||
375 | <listitem><para>Sets the machine name for this container. This | |
376 | name may be used to identify this container during its runtime | |
377 | (for example in tools like | |
378 | <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry> | |
379 | and similar), and is used to initialize the container's | |
380 | hostname (which the container can choose to override, | |
381 | however). If not specified, the last component of the root | |
382 | directory path of the container is used, possibly suffixed | |
383 | with a random identifier in case <option>--ephemeral</option> | |
384 | mode is selected. If the root directory selected is the host's | |
385 | root directory the host's hostname is used as default | |
386 | instead.</para></listitem> | |
387 | </varlistentry> | |
388 | ||
389 | <varlistentry> | |
390 | <term><option>--uuid=</option></term> | |
391 | ||
392 | <listitem><para>Set the specified UUID for the container. The | |
393 | init system will initialize | |
394 | <filename>/etc/machine-id</filename> from this if this file is | |
e01ff70a MS |
395 | not set yet. Note that this option takes effect only if |
396 | <filename>/etc/machine-id</filename> in the container is | |
397 | unpopulated.</para></listitem> | |
798d3a52 ZJS |
398 | </varlistentry> |
399 | ||
400 | <varlistentry> | |
401 | <term><option>--slice=</option></term> | |
402 | ||
cd2dfc6f LP |
403 | <listitem><para>Make the container part of the specified slice, instead of the default |
404 | <filename>machine.slice</filename>. This applies only if the machine is run in its own scope unit, i.e. if | |
405 | <option>--keep-unit</option> isn't used.</para> | |
f36933fe LP |
406 | </listitem> |
407 | </varlistentry> | |
408 | ||
409 | <varlistentry> | |
410 | <term><option>--property=</option></term> | |
411 | ||
cd2dfc6f LP |
412 | <listitem><para>Set a unit property on the scope unit to register for the machine. This applies only if the |
413 | machine is run in its own scope unit, i.e. if <option>--keep-unit</option> isn't used. Takes unit property | |
414 | assignments in the same format as <command>systemctl set-property</command>. This is useful to set memory | |
415 | limits and similar for container.</para> | |
798d3a52 ZJS |
416 | </listitem> |
417 | </varlistentry> | |
418 | ||
03cfe0d5 LP |
419 | <varlistentry> |
420 | <term><option>--private-users=</option></term> | |
421 | ||
d2e5535f LP |
422 | <listitem><para>Controls user namespacing. If enabled, the container will run with its own private set of UNIX |
423 | user and group ids (UIDs and GIDs). This involves mapping the private UIDs/GIDs used in the container (starting | |
424 | with the container's root user 0 and up) to a range of UIDs/GIDs on the host that are not used for other | |
425 | purposes (usually in the range beyond the host's UID/GID 65536). The parameter may be specified as follows:</para> | |
426 | ||
427 | <orderedlist> | |
2dd67817 | 428 | <listitem><para>If one or two colon-separated numbers are specified, user namespacing is turned on. The first |
ae209204 ZJS |
429 | parameter specifies the first host UID/GID to assign to the container, the second parameter specifies the |
430 | number of host UIDs/GIDs to assign to the container. If the second parameter is omitted, 65536 UIDs/GIDs are | |
431 | assigned.</para></listitem> | |
432 | ||
433 | <listitem><para>If the parameter is omitted, or true, user namespacing is turned on. The UID/GID range to | |
434 | use is determined automatically from the file ownership of the root directory of the container's directory | |
435 | tree. To use this option, make sure to prepare the directory tree in advance, and ensure that all files and | |
436 | directories in it are owned by UIDs/GIDs in the range you'd like to use. Also, make sure that used file ACLs | |
437 | exclusively reference UIDs/GIDs in the appropriate range. If this mode is used the number of UIDs/GIDs | |
438 | assigned to the container for use is 65536, and the UID/GID of the root directory must be a multiple of | |
439 | 65536.</para></listitem> | |
440 | ||
441 | <listitem><para>If the parameter is false, user namespacing is turned off. This is the default.</para> | |
442 | </listitem> | |
443 | ||
444 | <listitem><para>The special value <literal>pick</literal> turns on user namespacing. In this case the UID/GID | |
445 | range is automatically chosen. As first step, the file owner of the root directory of the container's | |
446 | directory tree is read, and it is checked that it is currently not used by the system otherwise (in | |
447 | particular, that no other container is using it). If this check is successful, the UID/GID range determined | |
2dd67817 | 448 | this way is used, similar to the behavior if "yes" is specified. If the check is not successful (and thus |
ae209204 ZJS |
449 | the UID/GID range indicated in the root directory's file owner is already used elsewhere) a new – currently |
450 | unused – UID/GID range of 65536 UIDs/GIDs is randomly chosen between the host UID/GIDs of 524288 and | |
451 | 1878982656, always starting at a multiple of 65536. This setting implies | |
452 | <option>--private-users-chown</option> (see below), which has the effect that the files and directories in | |
453 | the container's directory tree will be owned by the appropriate users of the range picked. Using this option | |
2dd67817 | 454 | makes user namespace behavior fully automatic. Note that the first invocation of a previously unused |
ae209204 ZJS |
455 | container image might result in picking a new UID/GID range for it, and thus in the (possibly expensive) file |
456 | ownership adjustment operation. However, subsequent invocations of the container will be cheap (unless of | |
457 | course the picked UID/GID range is assigned to a different use by then).</para></listitem> | |
d2e5535f LP |
458 | </orderedlist> |
459 | ||
460 | <para>It is recommended to assign at least 65536 UIDs/GIDs to each container, so that the usable UID/GID range in the | |
461 | container covers 16 bit. For best security, do not assign overlapping UID/GID ranges to multiple containers. It is | |
462 | hence a good idea to use the upper 16 bit of the host 32-bit UIDs/GIDs as container identifier, while the lower 16 | |
2dd67817 | 463 | bit encode the container UID/GID used. This is in fact the behavior enforced by the |
d2e5535f LP |
464 | <option>--private-users=pick</option> option.</para> |
465 | ||
466 | <para>When user namespaces are used, the GID range assigned to each container is always chosen identical to the | |
467 | UID range.</para> | |
468 | ||
469 | <para>In most cases, using <option>--private-users=pick</option> is the recommended option as it enhances | |
470 | container security massively and operates fully automatically in most cases.</para> | |
471 | ||
472 | <para>Note that the picked UID/GID range is not written to <filename>/etc/passwd</filename> or | |
473 | <filename>/etc/group</filename>. In fact, the allocation of the range is not stored persistently anywhere, | |
aa10469e LP |
474 | except in the file ownership of the files and directories of the container.</para> |
475 | ||
476 | <para>Note that when user namespacing is used file ownership on disk reflects this, and all of the container's | |
477 | files and directories are owned by the container's effective user and group IDs. This means that copying files | |
478 | from and to the container image requires correction of the numeric UID/GID values, according to the UID/GID | |
479 | shift applied.</para></listitem> | |
03cfe0d5 LP |
480 | </varlistentry> |
481 | ||
d2e5535f LP |
482 | <varlistentry> |
483 | <term><option>--private-users-chown</option></term> | |
484 | ||
485 | <listitem><para>If specified, all files and directories in the container's directory tree will adjusted so that | |
486 | they are owned to the appropriate UIDs/GIDs selected for the container (see above). This operation is | |
487 | potentially expensive, as it involves descending and iterating through the full directory tree of the | |
488 | container. Besides actual file ownership, file ACLs are adjusted as well.</para> | |
489 | ||
490 | <para>This option is implied if <option>--private-users=pick</option> is used. This option has no effect if | |
491 | user namespacing is not used.</para></listitem> | |
492 | </varlistentry> | |
03cfe0d5 | 493 | |
6265bde2 ZJS |
494 | <varlistentry> |
495 | <term><option>-U</option></term> | |
496 | ||
497 | <listitem><para>If the kernel supports the user namespaces feature, equivalent to | |
498 | <option>--private-users=pick --private-users-chown</option>, otherwise equivalent to | |
499 | <option>--private-users=no</option>.</para> | |
500 | ||
501 | <para>Note that <option>-U</option> is the default if the | |
502 | <filename>systemd-nspawn@.service</filename> template unit file is used.</para> | |
503 | ||
504 | <para>Note: it is possible to undo the effect of <option>--private-users-chown</option> (or | |
505 | <option>-U</option>) on the file system by redoing the operation with the first UID of 0:</para> | |
506 | ||
507 | <programlisting>systemd-nspawn … --private-users=0 --private-users-chown</programlisting> | |
508 | </listitem> | |
509 | </varlistentry> | |
510 | ||
798d3a52 ZJS |
511 | <varlistentry> |
512 | <term><option>--private-network</option></term> | |
513 | ||
514 | <listitem><para>Disconnect networking of the container from | |
515 | the host. This makes all network interfaces unavailable in the | |
516 | container, with the exception of the loopback device and those | |
517 | specified with <option>--network-interface=</option> and | |
518 | configured with <option>--network-veth</option>. If this | |
519 | option is specified, the CAP_NET_ADMIN capability will be | |
520 | added to the set of capabilities the container retains. The | |
521 | latter may be disabled by using | |
522 | <option>--drop-capability=</option>.</para></listitem> | |
523 | </varlistentry> | |
524 | ||
525 | <varlistentry> | |
526 | <term><option>--network-interface=</option></term> | |
527 | ||
528 | <listitem><para>Assign the specified network interface to the | |
529 | container. This will remove the specified interface from the | |
530 | calling namespace and place it in the container. When the | |
531 | container terminates, it is moved back to the host namespace. | |
532 | Note that <option>--network-interface=</option> implies | |
533 | <option>--private-network</option>. This option may be used | |
534 | more than once to add multiple network interfaces to the | |
535 | container.</para></listitem> | |
536 | </varlistentry> | |
537 | ||
538 | <varlistentry> | |
539 | <term><option>--network-macvlan=</option></term> | |
540 | ||
541 | <listitem><para>Create a <literal>macvlan</literal> interface | |
542 | of the specified Ethernet network interface and add it to the | |
543 | container. A <literal>macvlan</literal> interface is a virtual | |
544 | interface that adds a second MAC address to an existing | |
545 | physical Ethernet link. The interface in the container will be | |
546 | named after the interface on the host, prefixed with | |
547 | <literal>mv-</literal>. Note that | |
548 | <option>--network-macvlan=</option> implies | |
549 | <option>--private-network</option>. This option may be used | |
550 | more than once to add multiple network interfaces to the | |
551 | container.</para></listitem> | |
552 | </varlistentry> | |
553 | ||
554 | <varlistentry> | |
555 | <term><option>--network-ipvlan=</option></term> | |
556 | ||
557 | <listitem><para>Create an <literal>ipvlan</literal> interface | |
558 | of the specified Ethernet network interface and add it to the | |
559 | container. An <literal>ipvlan</literal> interface is a virtual | |
560 | interface, similar to a <literal>macvlan</literal> interface, | |
561 | which uses the same MAC address as the underlying interface. | |
562 | The interface in the container will be named after the | |
563 | interface on the host, prefixed with <literal>iv-</literal>. | |
564 | Note that <option>--network-ipvlan=</option> implies | |
565 | <option>--private-network</option>. This option may be used | |
566 | more than once to add multiple network interfaces to the | |
567 | container.</para></listitem> | |
568 | </varlistentry> | |
569 | ||
570 | <varlistentry> | |
571 | <term><option>-n</option></term> | |
572 | <term><option>--network-veth</option></term> | |
573 | ||
5e7423ff LP |
574 | <listitem><para>Create a virtual Ethernet link (<literal>veth</literal>) between host and container. The host |
575 | side of the Ethernet link will be available as a network interface named after the container's name (as | |
576 | specified with <option>--machine=</option>), prefixed with <literal>ve-</literal>. The container side of the | |
577 | Ethernet link will be named <literal>host0</literal>. The <option>--network-veth</option> option implies | |
578 | <option>--private-network</option>.</para> | |
579 | ||
580 | <para>Note that | |
581 | <citerefentry><refentrytitle>systemd-networkd.service</refentrytitle><manvolnum>8</manvolnum></citerefentry> | |
582 | includes by default a network file <filename>/usr/lib/systemd/network/80-container-ve.network</filename> | |
583 | matching the host-side interfaces created this way, which contains settings to enable automatic address | |
584 | provisioning on the created virtual link via DHCP, as well as automatic IP routing onto the host's external | |
585 | network interfaces. It also contains <filename>/usr/lib/systemd/network/80-container-host0.network</filename> | |
586 | matching the container-side interface created this way, containing settings to enable client side address | |
587 | assignment via DHCP. In case <filename>systemd-networkd</filename> is running on both the host and inside the | |
588 | container, automatic IP communication from the container to the host is thus available, with further | |
589 | connectivity to the external network.</para> | |
b09c0bba LP |
590 | |
591 | <para>Note that <option>--network-veth</option> is the default if the | |
592 | <filename>systemd-nspawn@.service</filename> template unit file is used.</para> | |
5e7423ff | 593 | </listitem> |
798d3a52 ZJS |
594 | </varlistentry> |
595 | ||
f6d6bad1 LP |
596 | <varlistentry> |
597 | <term><option>--network-veth-extra=</option></term> | |
598 | ||
599 | <listitem><para>Adds an additional virtual Ethernet link | |
600 | between host and container. Takes a colon-separated pair of | |
601 | host interface name and container interface name. The latter | |
602 | may be omitted in which case the container and host sides will | |
603 | be assigned the same name. This switch is independent of | |
ccddd104 | 604 | <option>--network-veth</option>, and — in contrast — may be |
f6d6bad1 LP |
605 | used multiple times, and allows configuration of the network |
606 | interface names. Note that <option>--network-bridge=</option> | |
607 | has no effect on interfaces created with | |
608 | <option>--network-veth-extra=</option>.</para></listitem> | |
609 | </varlistentry> | |
610 | ||
798d3a52 ZJS |
611 | <varlistentry> |
612 | <term><option>--network-bridge=</option></term> | |
613 | ||
5e7423ff LP |
614 | <listitem><para>Adds the host side of the Ethernet link created with <option>--network-veth</option> to the |
615 | specified Ethernet bridge interface. Expects a valid network interface name of a bridge device as | |
616 | argument. Note that <option>--network-bridge=</option> implies <option>--network-veth</option>. If this option | |
617 | is used, the host side of the Ethernet link will use the <literal>vb-</literal> prefix instead of | |
798d3a52 ZJS |
618 | <literal>ve-</literal>.</para></listitem> |
619 | </varlistentry> | |
620 | ||
938d2579 LP |
621 | <varlistentry> |
622 | <term><option>--network-zone=</option></term> | |
623 | ||
624 | <listitem><para>Creates a virtual Ethernet link (<literal>veth</literal>) to the container and adds it to an | |
625 | automatically managed Ethernet bridge interface. The bridge interface is named after the passed argument, | |
626 | prefixed with <literal>vz-</literal>. The bridge interface is automatically created when the first container | |
627 | configured for its name is started, and is automatically removed when the last container configured for its | |
628 | name exits. Hence, each bridge interface configured this way exists only as long as there's at least one | |
629 | container referencing it running. This option is very similar to <option>--network-bridge=</option>, besides | |
630 | this automatic creation/removal of the bridge device.</para> | |
631 | ||
632 | <para>This setting makes it easy to place multiple related containers on a common, virtual Ethernet-based | |
633 | broadcast domain, here called a "zone". Each container may only be part of one zone, but each zone may contain | |
634 | any number of containers. Each zone is referenced by its name. Names may be chosen freely (as long as they form | |
635 | valid network interface names when prefixed with <literal>vz-</literal>), and it is sufficient to pass the same | |
cf917c27 | 636 | name to the <option>--network-zone=</option> switch of the various concurrently running containers to join |
938d2579 LP |
637 | them in one zone.</para> |
638 | ||
639 | <para>Note that | |
640 | <citerefentry><refentrytitle>systemd-networkd.service</refentrytitle><manvolnum>8</manvolnum></citerefentry> | |
641 | includes by default a network file <filename>/usr/lib/systemd/network/80-container-vz.network</filename> | |
642 | matching the bridge interfaces created this way, which contains settings to enable automatic address | |
643 | provisioning on the created virtual network via DHCP, as well as automatic IP routing onto the host's external | |
644 | network interfaces. Using <option>--network-zone=</option> is hence in most cases fully automatic and | |
645 | sufficient to connect multiple local containers in a joined broadcast domain to the host, with further | |
646 | connectivity to the external network.</para> | |
647 | </listitem> | |
648 | </varlistentry> | |
649 | ||
798d3a52 ZJS |
650 | <varlistentry> |
651 | <term><option>-p</option></term> | |
652 | <term><option>--port=</option></term> | |
653 | ||
654 | <listitem><para>If private networking is enabled, maps an IP | |
655 | port on the host onto an IP port on the container. Takes a | |
656 | protocol specifier (either <literal>tcp</literal> or | |
657 | <literal>udp</literal>), separated by a colon from a host port | |
658 | number in the range 1 to 65535, separated by a colon from a | |
659 | container port number in the range from 1 to 65535. The | |
660 | protocol specifier and its separating colon may be omitted, in | |
661 | which case <literal>tcp</literal> is assumed. The container | |
7c918141 | 662 | port number and its colon may be omitted, in which case the |
798d3a52 | 663 | same port as the host port is implied. This option is only |
a8eaaee7 | 664 | supported if private networking is used, such as with |
938d2579 | 665 | <option>--network-veth</option>, <option>--network-zone=</option> |
798d3a52 ZJS |
666 | <option>--network-bridge=</option>.</para></listitem> |
667 | </varlistentry> | |
668 | ||
669 | <varlistentry> | |
670 | <term><option>-Z</option></term> | |
671 | <term><option>--selinux-context=</option></term> | |
672 | ||
673 | <listitem><para>Sets the SELinux security context to be used | |
674 | to label processes in the container.</para> | |
675 | </listitem> | |
676 | </varlistentry> | |
677 | ||
678 | <varlistentry> | |
679 | <term><option>-L</option></term> | |
680 | <term><option>--selinux-apifs-context=</option></term> | |
681 | ||
682 | <listitem><para>Sets the SELinux security context to be used | |
683 | to label files in the virtual API file systems in the | |
684 | container.</para> | |
685 | </listitem> | |
686 | </varlistentry> | |
687 | ||
688 | <varlistentry> | |
689 | <term><option>--capability=</option></term> | |
690 | ||
691 | <listitem><para>List one or more additional capabilities to | |
692 | grant the container. Takes a comma-separated list of | |
693 | capability names, see | |
694 | <citerefentry project='man-pages'><refentrytitle>capabilities</refentrytitle><manvolnum>7</manvolnum></citerefentry> | |
695 | for more information. Note that the following capabilities | |
696 | will be granted in any way: CAP_CHOWN, CAP_DAC_OVERRIDE, | |
697 | CAP_DAC_READ_SEARCH, CAP_FOWNER, CAP_FSETID, CAP_IPC_OWNER, | |
698 | CAP_KILL, CAP_LEASE, CAP_LINUX_IMMUTABLE, | |
699 | CAP_NET_BIND_SERVICE, CAP_NET_BROADCAST, CAP_NET_RAW, | |
700 | CAP_SETGID, CAP_SETFCAP, CAP_SETPCAP, CAP_SETUID, | |
701 | CAP_SYS_ADMIN, CAP_SYS_CHROOT, CAP_SYS_NICE, CAP_SYS_PTRACE, | |
702 | CAP_SYS_TTY_CONFIG, CAP_SYS_RESOURCE, CAP_SYS_BOOT, | |
703 | CAP_AUDIT_WRITE, CAP_AUDIT_CONTROL. Also CAP_NET_ADMIN is | |
704 | retained if <option>--private-network</option> is specified. | |
705 | If the special value <literal>all</literal> is passed, all | |
706 | capabilities are retained.</para></listitem> | |
707 | </varlistentry> | |
708 | ||
709 | <varlistentry> | |
710 | <term><option>--drop-capability=</option></term> | |
711 | ||
712 | <listitem><para>Specify one or more additional capabilities to | |
713 | drop for the container. This allows running the container with | |
714 | fewer capabilities than the default (see | |
715 | above).</para></listitem> | |
716 | </varlistentry> | |
717 | ||
960e4569 LP |
718 | <varlistentry> |
719 | <term><option>--system-call-filter=</option></term> | |
720 | ||
721 | <listitem><para>Alter the system call filter applied to containers. Takes a space-separated list of system call | |
722 | names or group names (the latter prefixed with <literal>@</literal>, as listed by the | |
723 | <command>syscall-filter</command> command of <citerefentry | |
724 | project='man-pages'><refentrytitle>systemd-analyze</refentrytitle><manvolnum>1</manvolnum></citerefentry>). Passed | |
725 | system calls will be permitted. The list may optionally be prefixed by <literal>~</literal>, in which case all | |
726 | listed system calls are prohibited. If this command line option is used multiple times the configured lists are | |
727 | combined. If both a positive and a negative list (that is one system call list without and one with the | |
96bedbe2 LP |
728 | <literal>~</literal> prefix) are configured, the negative list takes precedence over the positive list. Note |
729 | that <command>systemd-nspawn</command> always implements a system call whitelist (as opposed to a blacklist), | |
730 | and this command line option hence adds or removes entries from the default whitelist, depending on the | |
960e4569 LP |
731 | <literal>~</literal> prefix. Note that the applied system call filter is also altered implicitly if additional |
732 | capabilities are passed using the <command>--capabilities=</command>.</para></listitem> | |
733 | </varlistentry> | |
734 | ||
c6c8f6e2 LP |
735 | <varlistentry> |
736 | <term><option>--kill-signal=</option></term> | |
737 | ||
738 | <listitem><para>Specify the process signal to send to the | |
739 | container's PID 1 when nspawn itself receives SIGTERM, in | |
740 | order to trigger an orderly shutdown of the | |
741 | container. Defaults to SIGRTMIN+3 if <option>--boot</option> | |
742 | is used (on systemd-compatible init systems SIGRTMIN+3 | |
b3969f73 PA |
743 | triggers an orderly shutdown). For a list of valid signals, see |
744 | <citerefentry project='man-pages'><refentrytitle>signal</refentrytitle><manvolnum>7</manvolnum></citerefentry>.</para></listitem> | |
c6c8f6e2 LP |
745 | </varlistentry> |
746 | ||
798d3a52 ZJS |
747 | <varlistentry> |
748 | <term><option>--link-journal=</option></term> | |
749 | ||
750 | <listitem><para>Control whether the container's journal shall | |
751 | be made visible to the host system. If enabled, allows viewing | |
752 | the container's journal files from the host (but not vice | |
753 | versa). Takes one of <literal>no</literal>, | |
754 | <literal>host</literal>, <literal>try-host</literal>, | |
755 | <literal>guest</literal>, <literal>try-guest</literal>, | |
756 | <literal>auto</literal>. If <literal>no</literal>, the journal | |
757 | is not linked. If <literal>host</literal>, the journal files | |
758 | are stored on the host file system (beneath | |
759 | <filename>/var/log/journal/<replaceable>machine-id</replaceable></filename>) | |
760 | and the subdirectory is bind-mounted into the container at the | |
761 | same location. If <literal>guest</literal>, the journal files | |
762 | are stored on the guest file system (beneath | |
763 | <filename>/var/log/journal/<replaceable>machine-id</replaceable></filename>) | |
764 | and the subdirectory is symlinked into the host at the same | |
765 | location. <literal>try-host</literal> and | |
766 | <literal>try-guest</literal> do the same but do not fail if | |
2dd67817 | 767 | the host does not have persistent journaling enabled. If |
798d3a52 ZJS |
768 | <literal>auto</literal> (the default), and the right |
769 | subdirectory of <filename>/var/log/journal</filename> exists, | |
770 | it will be bind mounted into the container. If the | |
771 | subdirectory does not exist, no linking is performed. | |
772 | Effectively, booting a container once with | |
773 | <literal>guest</literal> or <literal>host</literal> will link | |
774 | the journal persistently if further on the default of | |
b09c0bba LP |
775 | <literal>auto</literal> is used.</para> |
776 | ||
777 | <para>Note that <option>--link-journal=try-guest</option> is the default if the | |
778 | <filename>systemd-nspawn@.service</filename> template unit file is used.</para></listitem> | |
798d3a52 ZJS |
779 | </varlistentry> |
780 | ||
781 | <varlistentry> | |
782 | <term><option>-j</option></term> | |
783 | ||
784 | <listitem><para>Equivalent to | |
785 | <option>--link-journal=try-guest</option>.</para></listitem> | |
786 | </varlistentry> | |
787 | ||
788 | <varlistentry> | |
789 | <term><option>--read-only</option></term> | |
790 | ||
791 | <listitem><para>Mount the root file system read-only for the | |
792 | container.</para></listitem> | |
793 | </varlistentry> | |
794 | ||
795 | <varlistentry> | |
796 | <term><option>--bind=</option></term> | |
797 | <term><option>--bind-ro=</option></term> | |
798 | ||
86c0dd4a | 799 | <listitem><para>Bind mount a file or directory from the host into the container. Takes one of: a path |
c7a4890c LP |
800 | argument — in which case the specified path will be mounted from the host to the same path in the container, or |
801 | a colon-separated pair of paths — in which case the first specified path is the source in the host, and the | |
802 | second path is the destination in the container, or a colon-separated triple of source path, destination path | |
86c0dd4a | 803 | and mount options. The source path may optionally be prefixed with a <literal>+</literal> character. If so, the |
c7a4890c LP |
804 | source path is taken relative to the image's root directory. This permits setting up bind mounts within the |
805 | container image. The source path may be specified as empty string, in which case a temporary directory below | |
806 | the host's <filename>/var/tmp</filename> directory is used. It is automatically removed when the container is | |
807 | shut down. Mount options are comma-separated and currently, only <option>rbind</option> and | |
808 | <option>norbind</option> are allowed, controlling whether to create a recursive or a regular bind | |
809 | mount. Defaults to "rbind". Backslash escapes are interpreted, so <literal>\:</literal> may be used to embed | |
810 | colons in either path. This option may be specified multiple times for creating multiple independent bind | |
994a6364 LP |
811 | mount points. The <option>--bind-ro=</option> option creates read-only bind mounts.</para> |
812 | ||
813 | <para>Note that when this option is used in combination with <option>--private-users</option>, the resulting | |
814 | mount points will be owned by the <constant>nobody</constant> user. That's because the mount and its files and | |
815 | directories continue to be owned by the relevant host users and groups, which do not exist in the container, | |
816 | and thus show up under the wildcard UID 65534 (nobody). If such bind mounts are created, it is recommended to | |
817 | make them read-only, using <option>--bind-ro=</option>.</para></listitem> | |
798d3a52 ZJS |
818 | </varlistentry> |
819 | ||
820 | <varlistentry> | |
821 | <term><option>--tmpfs=</option></term> | |
822 | ||
823 | <listitem><para>Mount a tmpfs file system into the container. | |
824 | Takes a single absolute path argument that specifies where to | |
825 | mount the tmpfs instance to (in which case the directory | |
826 | access mode will be chosen as 0755, owned by root/root), or | |
827 | optionally a colon-separated pair of path and mount option | |
b938cb90 | 828 | string that is used for mounting (in which case the kernel |
798d3a52 ZJS |
829 | default for access mode and owner will be chosen, unless |
830 | otherwise specified). This option is particularly useful for | |
831 | mounting directories such as <filename>/var</filename> as | |
832 | tmpfs, to allow state-less systems, in particular when | |
ffcd3e89 | 833 | combined with <option>--read-only</option>. |
b938cb90 | 834 | Backslash escapes are interpreted in the path, so |
ffcd3e89 RM |
835 | <literal>\:</literal> may be used to embed colons in the path. |
836 | </para></listitem> | |
798d3a52 ZJS |
837 | </varlistentry> |
838 | ||
5a8af538 LP |
839 | <varlistentry> |
840 | <term><option>--overlay=</option></term> | |
841 | <term><option>--overlay-ro=</option></term> | |
842 | ||
843 | <listitem><para>Combine multiple directory trees into one | |
844 | overlay file system and mount it into the container. Takes a | |
845 | list of colon-separated paths to the directory trees to | |
846 | combine and the destination mount point.</para> | |
847 | ||
2eadf91c RM |
848 | <para>Backslash escapes are interpreted in the paths, so |
849 | <literal>\:</literal> may be used to embed colons in the paths. | |
850 | </para> | |
851 | ||
5a8af538 LP |
852 | <para>If three or more paths are specified, then the last |
853 | specified path is the destination mount point in the | |
854 | container, all paths specified before refer to directory trees | |
855 | on the host and are combined in the specified order into one | |
856 | overlay file system. The left-most path is hence the lowest | |
857 | directory tree, the second-to-last path the highest directory | |
858 | tree in the stacking order. If <option>--overlay-ro=</option> | |
b938cb90 | 859 | is used instead of <option>--overlay=</option>, a read-only |
5a8af538 | 860 | overlay file system is created. If a writable overlay file |
b938cb90 | 861 | system is created, all changes made to it are written to the |
5a8af538 LP |
862 | highest directory tree in the stacking order, i.e. the |
863 | second-to-last specified.</para> | |
864 | ||
865 | <para>If only two paths are specified, then the second | |
866 | specified path is used both as the top-level directory tree in | |
867 | the stacking order as seen from the host, as well as the mount | |
868 | point for the overlay file system in the container. At least | |
869 | two paths have to be specified.</para> | |
870 | ||
86c0dd4a | 871 | <para>The source paths may optionally be prefixed with <literal>+</literal> character. If so they are taken |
c7a4890c LP |
872 | relative to the image's root directory. The uppermost source path may also be specified as empty string, in |
873 | which case a temporary directory below the host's <filename>/var/tmp</filename> is used. The directory is | |
874 | removed automatically when the container is shut down. This behaviour is useful in order to make read-only | |
875 | container directories writable while the container is running. For example, use the | |
876 | <literal>--overlay=+/var::/var</literal> option in order to automatically overlay a writable temporary | |
877 | directory on a read-only <filename>/var</filename> directory.</para> | |
86c0dd4a | 878 | |
5a8af538 LP |
879 | <para>For details about overlay file systems, see <ulink |
880 | url="https://www.kernel.org/doc/Documentation/filesystems/overlayfs.txt">overlayfs.txt</ulink>. Note | |
881 | that the semantics of overlay file systems are substantially | |
882 | different from normal file systems, in particular regarding | |
883 | reported device and inode information. Device and inode | |
884 | information may change for a file while it is being written | |
885 | to, and processes might see out-of-date versions of files at | |
886 | times. Note that this switch automatically derives the | |
887 | <literal>workdir=</literal> mount option for the overlay file | |
888 | system from the top-level directory tree, making it a sibling | |
889 | of it. It is hence essential that the top-level directory tree | |
890 | is not a mount point itself (since the working directory must | |
891 | be on the same file system as the top-most directory | |
892 | tree). Also note that the <literal>lowerdir=</literal> mount | |
893 | option receives the paths to stack in the opposite order of | |
894 | this switch.</para></listitem> | |
895 | </varlistentry> | |
896 | ||
798d3a52 | 897 | <varlistentry> |
a5f1cb3b ZJS |
898 | <term><option>-E <replaceable>NAME</replaceable>=<replaceable>VALUE</replaceable></option></term> |
899 | <term><option>--setenv=<replaceable>NAME</replaceable>=<replaceable>VALUE</replaceable></option></term> | |
798d3a52 ZJS |
900 | |
901 | <listitem><para>Specifies an environment variable assignment | |
902 | to pass to the init process in the container, in the format | |
903 | <literal>NAME=VALUE</literal>. This may be used to override | |
904 | the default variables or to set additional variables. This | |
905 | parameter may be used more than once.</para></listitem> | |
906 | </varlistentry> | |
907 | ||
798d3a52 ZJS |
908 | <varlistentry> |
909 | <term><option>--register=</option></term> | |
910 | ||
cd2dfc6f LP |
911 | <listitem><para>Controls whether the container is registered with |
912 | <citerefentry><refentrytitle>systemd-machined</refentrytitle><manvolnum>8</manvolnum></citerefentry>. Takes a | |
913 | boolean argument, which defaults to <literal>yes</literal>. This option should be enabled when the container | |
914 | runs a full Operating System (more specifically: a system and service manager as PID 1), and is useful to | |
915 | ensure that the container is accessible via | |
916 | <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry> and shown by | |
917 | tools such as <citerefentry | |
918 | project='man-pages'><refentrytitle>ps</refentrytitle><manvolnum>1</manvolnum></citerefentry>. If the container | |
919 | does not run a service manager, it is recommended to set this option to | |
920 | <literal>no</literal>.</para></listitem> | |
798d3a52 ZJS |
921 | </varlistentry> |
922 | ||
923 | <varlistentry> | |
924 | <term><option>--keep-unit</option></term> | |
925 | ||
bb849957 LP |
926 | <listitem><para>Instead of creating a transient scope unit to run the container in, simply use the service or |
927 | scope unit <command>systemd-nspawn</command> has been invoked in. If <option>--register=yes</option> is set | |
928 | this unit is registered with | |
929 | <citerefentry><refentrytitle>systemd-machined</refentrytitle><manvolnum>8</manvolnum></citerefentry>. This | |
930 | switch should be used if <command>systemd-nspawn</command> is invoked from within a service unit, and the | |
931 | service unit's sole purpose is to run a single <command>systemd-nspawn</command> container. This option is not | |
932 | available if run from a user session.</para> | |
cd2dfc6f | 933 | <para>Note that passing <option>--keep-unit</option> disables the effect of <option>--slice=</option> and |
bb849957 LP |
934 | <option>--property=</option>. Use <option>--keep-unit</option> and <option>--register=no</option> in |
935 | combination to disable any kind of unit allocation or registration with | |
936 | <command>systemd-machined</command>.</para></listitem> | |
798d3a52 ZJS |
937 | </varlistentry> |
938 | ||
939 | <varlistentry> | |
940 | <term><option>--personality=</option></term> | |
941 | ||
942 | <listitem><para>Control the architecture ("personality") | |
943 | reported by | |
3ba3a79d | 944 | <citerefentry project='man-pages'><refentrytitle>uname</refentrytitle><manvolnum>2</manvolnum></citerefentry> |
798d3a52 ZJS |
945 | in the container. Currently, only <literal>x86</literal> and |
946 | <literal>x86-64</literal> are supported. This is useful when | |
947 | running a 32-bit container on a 64-bit host. If this setting | |
948 | is not used, the personality reported in the container is the | |
949 | same as the one reported on the host.</para></listitem> | |
950 | </varlistentry> | |
951 | ||
952 | <varlistentry> | |
953 | <term><option>-q</option></term> | |
954 | <term><option>--quiet</option></term> | |
955 | ||
956 | <listitem><para>Turns off any status output by the tool | |
957 | itself. When this switch is used, the only output from nspawn | |
958 | will be the console output of the container OS | |
959 | itself.</para></listitem> | |
960 | </varlistentry> | |
961 | ||
962 | <varlistentry> | |
f757855e LP |
963 | <term><option>--volatile</option></term> |
964 | <term><option>--volatile=</option><replaceable>MODE</replaceable></term> | |
798d3a52 ZJS |
965 | |
966 | <listitem><para>Boots the container in volatile mode. When no | |
967 | mode parameter is passed or when mode is specified as | |
b938cb90 | 968 | <option>yes</option>, full volatile mode is enabled. This |
a8eaaee7 | 969 | means the root directory is mounted as a mostly unpopulated |
798d3a52 | 970 | <literal>tmpfs</literal> instance, and |
cd72d204 JE |
971 | <filename>/usr</filename> from the OS tree is mounted into it |
972 | in read-only mode (the system thus starts up with read-only OS | |
5164c3b4 ZJS |
973 | image, but pristine state and configuration, any changes |
974 | are lost on shutdown). When the mode parameter | |
b938cb90 | 975 | is specified as <option>state</option>, the OS tree is |
798d3a52 | 976 | mounted read-only, but <filename>/var</filename> is mounted as |
a8eaaee7 | 977 | a <literal>tmpfs</literal> instance into it (the system thus |
798d3a52 | 978 | starts up with read-only OS resources and configuration, but |
a8eaaee7 | 979 | pristine state, and any changes to the latter are lost on |
798d3a52 | 980 | shutdown). When the mode parameter is specified as |
b938cb90 | 981 | <option>no</option> (the default), the whole OS tree is made |
798d3a52 ZJS |
982 | available writable.</para> |
983 | ||
91214a37 LP |
984 | <para>This option provides similar functionality for containers as the <literal>systemd.volatile=</literal> |
985 | kernel command line switch provides for host systems. See | |
986 | <citerefentry><refentrytitle>kernel-command-line</refentrytitle><manvolnum>7</manvolnum></citerefentry> for | |
987 | details.</para> | |
988 | ||
989 | <para>Note that enabling this setting will only work correctly with operating systems in the container that can | |
990 | boot up with only <filename>/usr</filename> mounted, and are able to automatically populate | |
991 | <filename>/var</filename>, and also <filename>/etc</filename> in case of | |
992 | <literal>--volatile=yes</literal>.</para></listitem> | |
798d3a52 ZJS |
993 | </varlistentry> |
994 | ||
f757855e LP |
995 | <varlistentry> |
996 | <term><option>--settings=</option><replaceable>MODE</replaceable></term> | |
997 | ||
998 | <listitem><para>Controls whether | |
999 | <command>systemd-nspawn</command> shall search for and use | |
1000 | additional per-container settings from | |
1001 | <filename>.nspawn</filename> files. Takes a boolean or the | |
1002 | special values <option>override</option> or | |
1003 | <option>trusted</option>.</para> | |
1004 | ||
b938cb90 | 1005 | <para>If enabled (the default), a settings file named after the |
f757855e LP |
1006 | machine (as specified with the <option>--machine=</option> |
1007 | setting, or derived from the directory or image file name) | |
1008 | with the suffix <filename>.nspawn</filename> is searched in | |
1009 | <filename>/etc/systemd/nspawn/</filename> and | |
1010 | <filename>/run/systemd/nspawn/</filename>. If it is found | |
1011 | there, its settings are read and used. If it is not found | |
b938cb90 | 1012 | there, it is subsequently searched in the same directory as the |
f757855e | 1013 | image file or in the immediate parent of the root directory of |
b938cb90 | 1014 | the container. In this case, if the file is found, its settings |
f757855e | 1015 | will be also read and used, but potentially unsafe settings |
b938cb90 | 1016 | are ignored. Note that in both these cases, settings on the |
4f76ef04 | 1017 | command line take precedence over the corresponding settings |
f757855e LP |
1018 | from loaded <filename>.nspawn</filename> files, if both are |
1019 | specified. Unsafe settings are considered all settings that | |
1020 | elevate the container's privileges or grant access to | |
1021 | additional resources such as files or directories of the | |
1022 | host. For details about the format and contents of | |
b938cb90 | 1023 | <filename>.nspawn</filename> files, consult |
f757855e LP |
1024 | <citerefentry><refentrytitle>systemd.nspawn</refentrytitle><manvolnum>5</manvolnum></citerefentry>.</para> |
1025 | ||
b938cb90 JE |
1026 | <para>If this option is set to <option>override</option>, the |
1027 | file is searched, read and used the same way, however, the order of | |
f757855e LP |
1028 | precedence is reversed: settings read from the |
1029 | <filename>.nspawn</filename> file will take precedence over | |
1030 | the corresponding command line options, if both are | |
1031 | specified.</para> | |
1032 | ||
b938cb90 | 1033 | <para>If this option is set to <option>trusted</option>, the |
f757855e | 1034 | file is searched, read and used the same way, but regardless |
a8eaaee7 | 1035 | of being found in <filename>/etc/systemd/nspawn/</filename>, |
f757855e LP |
1036 | <filename>/run/systemd/nspawn/</filename> or next to the image |
1037 | file or container root directory, all settings will take | |
b938cb90 | 1038 | effect, however, command line arguments still take precedence |
f757855e LP |
1039 | over corresponding settings.</para> |
1040 | ||
b938cb90 | 1041 | <para>If disabled, no <filename>.nspawn</filename> file is read |
f757855e LP |
1042 | and no settings except the ones on the command line are in |
1043 | effect.</para></listitem> | |
1044 | </varlistentry> | |
1045 | ||
9c1e04d0 | 1046 | <varlistentry> |
b09c0bba | 1047 | <term><option>--notify-ready=</option></term> |
9c1e04d0 AP |
1048 | |
1049 | <listitem><para>Configures support for notifications from the container's init process. | |
b09c0bba | 1050 | <option>--notify-ready=</option> takes a boolean (<option>no</option> and <option>yes</option>). |
9c1e04d0 AP |
1051 | With option <option>no</option> systemd-nspawn notifies systemd |
1052 | with a <literal>READY=1</literal> message when the init process is created. | |
1053 | With option <option>yes</option> systemd-nspawn waits for the | |
1054 | <literal>READY=1</literal> message from the init process in the container | |
1055 | before sending its own to systemd. For more details about notifications | |
1056 | see <citerefentry><refentrytitle>sd_notify</refentrytitle><manvolnum>3</manvolnum></citerefentry>).</para></listitem> | |
1057 | </varlistentry> | |
1058 | ||
798d3a52 ZJS |
1059 | <xi:include href="standard-options.xml" xpointer="help" /> |
1060 | <xi:include href="standard-options.xml" xpointer="version" /> | |
1061 | </variablelist> | |
1062 | ||
1063 | </refsect1> | |
1064 | ||
1065 | <refsect1> | |
1066 | <title>Examples</title> | |
1067 | ||
1068 | <example> | |
12c4ee0a ZJS |
1069 | <title>Download a |
1070 | <ulink url="https://getfedora.org">Fedora</ulink> image and start a shell in it</title> | |
798d3a52 | 1071 | |
3797fd0a ZJS |
1072 | <programlisting># machinectl pull-raw --verify=no \ |
1073 | https://download.fedoraproject.org/pub/fedora/linux/releases/25/CloudImages/x86_64/images/Fedora-Cloud-Base-25-1.3.x86_64.raw.xz | |
1074 | # systemd-nspawn -M Fedora-Cloud-Base-25-1.3.x86_64.raw</programlisting> | |
e0ea94c1 | 1075 | |
798d3a52 ZJS |
1076 | <para>This downloads an image using |
1077 | <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry> | |
1078 | and opens a shell in it.</para> | |
1079 | </example> | |
e0ea94c1 | 1080 | |
798d3a52 ZJS |
1081 | <example> |
1082 | <title>Build and boot a minimal Fedora distribution in a container</title> | |
8f7a3c14 | 1083 | |
55107232 | 1084 | <programlisting># dnf -y --releasever=27 --installroot=/var/lib/machines/f27container \ |
3797fd0a ZJS |
1085 | --disablerepo='*' --enablerepo=fedora --enablerepo=updates install \ |
1086 | systemd passwd dnf fedora-release vim-minimal | |
55107232 | 1087 | # systemd-nspawn -bD /var/lib/machines/f27container</programlisting> |
8f7a3c14 | 1088 | |
798d3a52 | 1089 | <para>This installs a minimal Fedora distribution into the |
55107232 ZJS |
1090 | directory <filename noindex='true'>/var/lib/machines/f27container</filename> |
1091 | and then boots an OS in a namespace container in it. Because the installation | |
1092 | is located underneath the standard <filename>/var/lib/machines/</filename> | |
1093 | directory, it is also possible to start the machine using | |
1094 | <command>systemd-nspawn -M f27container</command>.</para> | |
798d3a52 | 1095 | </example> |
8f7a3c14 | 1096 | |
798d3a52 ZJS |
1097 | <example> |
1098 | <title>Spawn a shell in a container of a minimal Debian unstable distribution</title> | |
8f7a3c14 | 1099 | |
798d3a52 | 1100 | <programlisting># debootstrap --arch=amd64 unstable ~/debian-tree/ |
25f5971b | 1101 | # systemd-nspawn -D ~/debian-tree/</programlisting> |
8f7a3c14 | 1102 | |
798d3a52 ZJS |
1103 | <para>This installs a minimal Debian unstable distribution into |
1104 | the directory <filename>~/debian-tree/</filename> and then | |
1105 | spawns a shell in a namespace container in it.</para> | |
12c4ee0a ZJS |
1106 | |
1107 | <para><command>debootstrap</command> supports | |
1108 | <ulink url="https://www.debian.org">Debian</ulink>, | |
1109 | <ulink url="https://www.ubuntu.com">Ubuntu</ulink>, | |
1110 | and <ulink url="https://www.tanglu.org">Tanglu</ulink> | |
1111 | out of the box, so the same command can be used to install any of those. For other | |
1112 | distributions from the Debian family, a mirror has to be specified, see | |
1113 | <citerefentry project='die-net'><refentrytitle>debootstrap</refentrytitle><manvolnum>8</manvolnum></citerefentry>. | |
1114 | </para> | |
798d3a52 | 1115 | </example> |
8f7a3c14 | 1116 | |
798d3a52 | 1117 | <example> |
12c4ee0a ZJS |
1118 | <title>Boot a minimal |
1119 | <ulink url="https://www.archlinux.org">Arch Linux</ulink> distribution in a container</title> | |
68562936 | 1120 | |
798d3a52 | 1121 | <programlisting># pacstrap -c -d ~/arch-tree/ base |
68562936 WG |
1122 | # systemd-nspawn -bD ~/arch-tree/</programlisting> |
1123 | ||
ff9b60f3 | 1124 | <para>This installs a minimal Arch Linux distribution into the |
798d3a52 ZJS |
1125 | directory <filename>~/arch-tree/</filename> and then boots an OS |
1126 | in a namespace container in it.</para> | |
1127 | </example> | |
68562936 | 1128 | |
f518ee04 | 1129 | <example> |
12c4ee0a ZJS |
1130 | <title>Install the |
1131 | <ulink url="https://software.opensuse.org/distributions/tumbleweed">OpenSUSE Tumbleweed</ulink> | |
1132 | rolling distribution</title> | |
f518ee04 ZJS |
1133 | |
1134 | <programlisting># zypper --root=/var/lib/machines/tumbleweed ar -c \ | |
1135 | https://download.opensuse.org/tumbleweed/repo/oss tumbleweed | |
1136 | # zypper --root=/var/lib/machines/tumbleweed refresh | |
1137 | # zypper --root=/var/lib/machines/tumbleweed install --no-recommends \ | |
1138 | systemd shadow zypper openSUSE-release vim | |
1139 | # systemd-nspawn -M tumbleweed passwd root | |
1140 | # systemd-nspawn -M tumbleweed -b</programlisting> | |
1141 | </example> | |
1142 | ||
798d3a52 | 1143 | <example> |
17cbb288 | 1144 | <title>Boot into an ephemeral snapshot of the host system</title> |
f9f4dd51 | 1145 | |
798d3a52 | 1146 | <programlisting># systemd-nspawn -D / -xb</programlisting> |
f9f4dd51 | 1147 | |
17cbb288 LP |
1148 | <para>This runs a copy of the host system in a snapshot which is removed immediately when the container |
1149 | exits. All file system changes made during runtime will be lost on shutdown, hence.</para> | |
798d3a52 | 1150 | </example> |
f9f4dd51 | 1151 | |
798d3a52 ZJS |
1152 | <example> |
1153 | <title>Run a container with SELinux sandbox security contexts</title> | |
a8828ed9 | 1154 | |
798d3a52 | 1155 | <programlisting># chcon system_u:object_r:svirt_sandbox_file_t:s0:c0,c1 -R /srv/container |
3797fd0a ZJS |
1156 | # systemd-nspawn -L system_u:object_r:svirt_sandbox_file_t:s0:c0,c1 \ |
1157 | -Z system_u:system_r:svirt_lxc_net_t:s0:c0,c1 -D /srv/container /bin/sh</programlisting> | |
798d3a52 | 1158 | </example> |
b53ede69 PW |
1159 | |
1160 | <example> | |
1161 | <title>Run a container with an OSTree deployment</title> | |
1162 | ||
3797fd0a ZJS |
1163 | <programlisting># systemd-nspawn -b -i ~/image.raw \ |
1164 | --pivot-root=/ostree/deploy/$OS/deploy/$CHECKSUM:/sysroot \ | |
1165 | --bind=+/sysroot/ostree/deploy/$OS/var:/var</programlisting> | |
b53ede69 | 1166 | </example> |
798d3a52 ZJS |
1167 | </refsect1> |
1168 | ||
1169 | <refsect1> | |
1170 | <title>Exit status</title> | |
1171 | ||
1172 | <para>The exit code of the program executed in the container is | |
1173 | returned.</para> | |
1174 | </refsect1> | |
1175 | ||
1176 | <refsect1> | |
1177 | <title>See Also</title> | |
1178 | <para> | |
1179 | <citerefentry><refentrytitle>systemd</refentrytitle><manvolnum>1</manvolnum></citerefentry>, | |
f757855e | 1180 | <citerefentry><refentrytitle>systemd.nspawn</refentrytitle><manvolnum>5</manvolnum></citerefentry>, |
798d3a52 ZJS |
1181 | <citerefentry project='man-pages'><refentrytitle>chroot</refentrytitle><manvolnum>1</manvolnum></citerefentry>, |
1182 | <citerefentry project='mankier'><refentrytitle>dnf</refentrytitle><manvolnum>8</manvolnum></citerefentry>, | |
798d3a52 ZJS |
1183 | <citerefentry project='die-net'><refentrytitle>debootstrap</refentrytitle><manvolnum>8</manvolnum></citerefentry>, |
1184 | <citerefentry project='archlinux'><refentrytitle>pacman</refentrytitle><manvolnum>8</manvolnum></citerefentry>, | |
f518ee04 | 1185 | <citerefentry project='mankier'><refentrytitle>zypper</refentrytitle><manvolnum>8</manvolnum></citerefentry>, |
798d3a52 ZJS |
1186 | <citerefentry><refentrytitle>systemd.slice</refentrytitle><manvolnum>5</manvolnum></citerefentry>, |
1187 | <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry>, | |
3ba3a79d | 1188 | <citerefentry project='man-pages'><refentrytitle>btrfs</refentrytitle><manvolnum>8</manvolnum></citerefentry> |
798d3a52 ZJS |
1189 | </para> |
1190 | </refsect1> | |
8f7a3c14 LP |
1191 | |
1192 | </refentry> |