]>
Commit | Line | Data |
---|---|---|
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 | ||
5 | <!-- | |
6 | This file is part of systemd. | |
7 | ||
8 | Copyright 2010 Lennart Poettering | |
9 | ||
10 | systemd is free software; you can redistribute it and/or modify it | |
11 | under the terms of the GNU Lesser General Public License as published by | |
12 | the Free Software Foundation; either version 2.1 of the License, or | |
13 | (at your option) any later version. | |
14 | ||
15 | systemd is distributed in the hope that it will be useful, but | |
16 | WITHOUT ANY WARRANTY; without even the implied warranty of | |
17 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | |
18 | Lesser General Public License for more details. | |
19 | ||
20 | You should have received a copy of the GNU Lesser General Public License | |
21 | along with systemd; If not, see <http://www.gnu.org/licenses/>. | |
22 | --> | |
23 | ||
24 | <refentry id="systemd-nspawn" | |
25 | xmlns:xi="http://www.w3.org/2001/XInclude"> | |
26 | ||
27 | <refentryinfo> | |
28 | <title>systemd-nspawn</title> | |
29 | <productname>systemd</productname> | |
30 | ||
31 | <authorgroup> | |
32 | <author> | |
33 | <contrib>Developer</contrib> | |
34 | <firstname>Lennart</firstname> | |
35 | <surname>Poettering</surname> | |
36 | <email>lennart@poettering.net</email> | |
37 | </author> | |
38 | </authorgroup> | |
39 | </refentryinfo> | |
40 | ||
41 | <refmeta> | |
42 | <refentrytitle>systemd-nspawn</refentrytitle> | |
43 | <manvolnum>1</manvolnum> | |
44 | </refmeta> | |
45 | ||
46 | <refnamediv> | |
47 | <refname>systemd-nspawn</refname> | |
48 | <refpurpose>Spawn a namespace container for debugging, testing and building</refpurpose> | |
49 | </refnamediv> | |
50 | ||
51 | <refsynopsisdiv> | |
52 | <cmdsynopsis> | |
53 | <command>systemd-nspawn</command> | |
54 | <arg choice="opt" rep="repeat">OPTIONS</arg> | |
55 | <arg choice="opt"><replaceable>COMMAND</replaceable> | |
56 | <arg choice="opt" rep="repeat">ARGS</arg> | |
57 | </arg> | |
58 | </cmdsynopsis> | |
59 | <cmdsynopsis> | |
60 | <command>systemd-nspawn</command> | |
61 | <arg choice="plain">--boot</arg> | |
62 | <arg choice="opt" rep="repeat">OPTIONS</arg> | |
63 | <arg choice="opt" rep="repeat">ARGS</arg> | |
64 | </cmdsynopsis> | |
65 | </refsynopsisdiv> | |
66 | ||
67 | <refsect1> | |
68 | <title>Description</title> | |
69 | ||
70 | <para><command>systemd-nspawn</command> may be used to run a command or OS in a light-weight namespace | |
71 | container. In many ways it is similar to <citerefentry | |
72 | project='man-pages'><refentrytitle>chroot</refentrytitle><manvolnum>1</manvolnum></citerefentry>, but more powerful | |
73 | since it fully virtualizes the file system hierarchy, as well as the process tree, the various IPC subsystems and | |
74 | the host and domain name.</para> | |
75 | ||
76 | <para><command>systemd-nspawn</command> may be invoked on any directory tree containing an operating system tree, | |
77 | using the <option>--directory=</option> command line option. By using the <option>--machine=</option> option an OS | |
78 | tree is automatically searched for in a couple of locations, most importantly in | |
79 | <filename>/var/lib/machines</filename>, the suggested directory to place container images installed on the | |
80 | system.</para> | |
81 | ||
82 | <para>In contrast to <citerefentry | |
83 | project='man-pages'><refentrytitle>chroot</refentrytitle><manvolnum>1</manvolnum></citerefentry> <command>systemd-nspawn</command> | |
84 | may be used to boot full Linux-based operating systems in a container.</para> | |
85 | ||
86 | <para><command>systemd-nspawn</command> limits access to various kernel interfaces in the container to read-only, | |
87 | such as <filename>/sys</filename>, <filename>/proc/sys</filename> or <filename>/sys/fs/selinux</filename>. The | |
88 | host's network interfaces and the system clock may not be changed from within the container. Device nodes may not | |
89 | be created. The host system cannot be rebooted and kernel modules may not be loaded from within the | |
90 | container.</para> | |
91 | ||
92 | <para>Use a tool like <citerefentry | |
93 | project='mankier'><refentrytitle>dnf</refentrytitle><manvolnum>8</manvolnum></citerefentry>, <citerefentry | |
94 | project='die-net'><refentrytitle>debootstrap</refentrytitle><manvolnum>8</manvolnum></citerefentry>, or | |
95 | <citerefentry project='archlinux'><refentrytitle>pacman</refentrytitle><manvolnum>8</manvolnum></citerefentry> to | |
96 | set up an OS directory tree suitable as file system hierarchy for <command>systemd-nspawn</command> containers. See | |
97 | the Examples section below for details on suitable invocation of these commands.</para> | |
98 | ||
99 | <para>As a safety check <command>systemd-nspawn</command> will verify the existence of | |
100 | <filename>/usr/lib/os-release</filename> or <filename>/etc/os-release</filename> in the container tree before | |
101 | starting the container (see | |
102 | <citerefentry><refentrytitle>os-release</refentrytitle><manvolnum>5</manvolnum></citerefentry>). It might be | |
103 | necessary to add this file to the container tree manually if the OS of the container is too old to contain this | |
104 | file out-of-the-box.</para> | |
105 | ||
106 | <para><command>systemd-nspawn</command> may be invoked directly from the interactive command line or run as system | |
107 | service in the background. In this mode each container instance runs as its own service instance; a default | |
108 | template unit file <filename>systemd-nspawn@.service</filename> is provided to make this easy, taking the container | |
109 | name as instance identifier. Note that different default options apply when <command>systemd-nspawn</command> is | |
110 | invoked by the template unit file than interactively on the command line. Most importantly the template unit file | |
111 | makes use of the <option>--boot</option> which is not the default in case <command>systemd-nspawn</command> is | |
112 | invoked from the interactive command line. Further differences with the defaults are documented along with the | |
113 | various supported options below.</para> | |
114 | ||
115 | <para>The <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry> tool may | |
116 | be used to execute a number of operations on containers. In particular it provides easy-to-use commands to run | |
117 | containers as system services using the <filename>systemd-nspawn@.service</filename> template unit | |
118 | file.</para> | |
119 | ||
120 | <para>Along with each container a settings file with the <filename>.nspawn</filename> suffix may exist, containing | |
121 | additional settings to apply when running the container. See | |
122 | <citerefentry><refentrytitle>systemd.nspawn</refentrytitle><manvolnum>5</manvolnum></citerefentry> for | |
123 | details. Settings files override the default options used by the <filename>systemd-nspawn@.service</filename> | |
124 | template unit file, making it usually unnecessary to alter this template file directly.</para> | |
125 | ||
126 | <para>Note that <command>systemd-nspawn</command> will mount file systems private to the container to | |
127 | <filename>/dev</filename>, <filename>/run</filename> and similar. These will not be visible outside of the | |
128 | container, and their contents will be lost when the container exits.</para> | |
129 | ||
130 | <para>Note that running two <command>systemd-nspawn</command> containers from the same directory tree will not make | |
131 | processes in them see each other. The PID namespace separation of the two containers is complete and the containers | |
132 | will share very few runtime objects except for the underlying file system. Use | |
133 | <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry>'s | |
134 | <command>login</command> or <command>shell</command> commands to request an additional login session in a running | |
135 | container.</para> | |
136 | ||
137 | <para><command>systemd-nspawn</command> implements the <ulink | |
138 | url="https://www.freedesktop.org/wiki/Software/systemd/ContainerInterface">Container Interface</ulink> | |
139 | specification.</para> | |
140 | ||
141 | <para>While running, containers invoked with <command>systemd-nspawn</command> are registered with the | |
142 | <citerefentry><refentrytitle>systemd-machined</refentrytitle><manvolnum>8</manvolnum></citerefentry> service that | |
143 | keeps track of running containers, and provides programming interfaces to interact with them.</para> | |
144 | </refsect1> | |
145 | ||
146 | <refsect1> | |
147 | <title>Options</title> | |
148 | ||
149 | <para>If option <option>-b</option> is specified, the arguments | |
150 | are used as arguments for the init binary. Otherwise, | |
151 | <replaceable>COMMAND</replaceable> specifies the program to launch | |
152 | in the container, and the remaining arguments are used as | |
153 | arguments for this program. If <option>--boot</option> is not used and | |
154 | no arguments are specified, a shell is launched in the | |
155 | container.</para> | |
156 | ||
157 | <para>The following options are understood:</para> | |
158 | ||
159 | <variablelist> | |
160 | <varlistentry> | |
161 | <term><option>-D</option></term> | |
162 | <term><option>--directory=</option></term> | |
163 | ||
164 | <listitem><para>Directory to use as file system root for the | |
165 | container.</para> | |
166 | ||
167 | <para>If neither <option>--directory=</option>, nor | |
168 | <option>--image=</option> is specified the directory is | |
169 | determined by searching for a directory named the same as the | |
170 | machine name specified with <option>--machine=</option>. See | |
171 | <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry> | |
172 | section "Files and Directories" for the precise search path.</para> | |
173 | ||
174 | <para>If neither <option>--directory=</option>, | |
175 | <option>--image=</option>, nor <option>--machine=</option> | |
176 | are specified, the current directory will | |
177 | be used. May not be specified together with | |
178 | <option>--image=</option>.</para></listitem> | |
179 | </varlistentry> | |
180 | ||
181 | <varlistentry> | |
182 | <term><option>--template=</option></term> | |
183 | ||
184 | <listitem><para>Directory or <literal>btrfs</literal> subvolume to use as template for the container's root | |
185 | directory. If this is specified and the container's root directory (as configured by | |
186 | <option>--directory=</option>) does not yet exist it is created as <literal>btrfs</literal> snapshot (if | |
187 | supported) or plain directory (otherwise) and populated from this template tree. Ideally, the specified | |
188 | template path refers to the root of a <literal>btrfs</literal> subvolume, in which case a simple copy-on-write | |
189 | snapshot is taken, and populating the root directory is instant. If the specified template path does not refer | |
190 | to the root of a <literal>btrfs</literal> subvolume (or not even to a <literal>btrfs</literal> file system at | |
191 | all), the tree is copied (though possibly in a copy-on-write scheme — if the file system supports that), which | |
192 | can be substantially more time-consuming. May not be specified together with <option>--image=</option> or | |
193 | <option>--ephemeral</option>.</para> | |
194 | ||
195 | <para>Note that this switch leaves host name, machine ID and | |
196 | all other settings that could identify the instance | |
197 | unmodified.</para></listitem> | |
198 | </varlistentry> | |
199 | ||
200 | <varlistentry> | |
201 | <term><option>-x</option></term> | |
202 | <term><option>--ephemeral</option></term> | |
203 | ||
204 | <listitem><para>If specified, the container is run with a temporary snapshot of its file system that is removed | |
205 | immediately when the container terminates. May not be specified together with | |
206 | <option>--template=</option>.</para> | |
207 | <para>Note that this switch leaves host name, machine ID and | |
208 | all other settings that could identify the instance | |
209 | unmodified.</para></listitem> | |
210 | </varlistentry> | |
211 | ||
212 | <varlistentry> | |
213 | <term><option>-i</option></term> | |
214 | <term><option>--image=</option></term> | |
215 | ||
216 | <listitem><para>Disk image to mount the root directory for the | |
217 | container from. Takes a path to a regular file or to a block | |
218 | device node. The file or block device must contain | |
219 | either:</para> | |
220 | ||
221 | <itemizedlist> | |
222 | <listitem><para>An MBR partition table with a single | |
223 | partition of type 0x83 that is marked | |
224 | bootable.</para></listitem> | |
225 | ||
226 | <listitem><para>A GUID partition table (GPT) with a single | |
227 | partition of type | |
228 | 0fc63daf-8483-4772-8e79-3d69d8477de4.</para></listitem> | |
229 | ||
230 | <listitem><para>A GUID partition table (GPT) with a marked | |
231 | root partition which is mounted as the root directory of the | |
232 | container. Optionally, GPT images may contain a home and/or | |
233 | a server data partition which are mounted to the appropriate | |
234 | places in the container. All these partitions must be | |
235 | identified by the partition types defined by the <ulink | |
236 | url="https://www.freedesktop.org/wiki/Specifications/DiscoverablePartitionsSpec/">Discoverable | |
237 | Partitions Specification</ulink>.</para></listitem> | |
238 | ||
239 | <listitem><para>No partition table, and a single file system spanning the whole image.</para></listitem> | |
240 | </itemizedlist> | |
241 | ||
242 | <para>On GPT images, if an EFI System Partition (ESP) is discovered, it is automatically mounted to | |
243 | <filename>/efi</filename> (or <filename>/boot</filename> as fallback) in case a directory by this name exists | |
244 | and is empty.</para> | |
245 | ||
246 | <para>Partitions encrypted with LUKS are automatically decrypted. Also, on GPT images dm-verity data integrity | |
247 | hash partitions are set up if the root hash for them is specified using the <option>--root-hash=</option> | |
248 | option.</para> | |
249 | ||
250 | <para>Any other partitions, such as foreign partitions or swap partitions are not mounted. May not be specified | |
251 | together with <option>--directory=</option>, <option>--template=</option>.</para></listitem> | |
252 | </varlistentry> | |
253 | ||
254 | <varlistentry> | |
255 | <term><option>--root-hash=</option></term> | |
256 | ||
257 | <listitem><para>Takes a data integrity (dm-verity) root hash specified in hexadecimal. This option enables data | |
258 | integrity checks using dm-verity, if the used image contains the appropriate integrity data (see above). The | |
259 | specified hash must match the root hash of integrity data, and is usually at least 256 bits (and hence 64 | |
260 | formatted hexadecimal characters) long (in case of SHA256 for example). If this option is not specified, but | |
261 | the image file carries the <literal>user.verity.roothash</literal> extended file attribute (see <citerefentry | |
262 | project='man-pages'><refentrytitle>xattr</refentrytitle><manvolnum>7</manvolnum></citerefentry>), then the root | |
263 | hash is read from it, also as formatted hexadecimal characters. If the extended file attribute is not found (or | |
264 | is not supported by the underlying file system), but a file with the <filename>.roothash</filename> suffix is | |
265 | found next to the image file, bearing otherwise the same name, the root hash is read from it and automatically | |
266 | used, also as formatted hexadecimal characters.</para></listitem> | |
267 | </varlistentry> | |
268 | ||
269 | <varlistentry> | |
270 | <term><option>-a</option></term> | |
271 | <term><option>--as-pid2</option></term> | |
272 | ||
273 | <listitem><para>Invoke the shell or specified program as process ID (PID) 2 instead of PID 1 (init). By | |
274 | default, if neither this option nor <option>--boot</option> is used, the selected binary is run as process with | |
275 | PID 1, a mode only suitable for programs that are aware of the special semantics that the process with PID 1 | |
276 | has on UNIX. For example, it needs to reap all processes reparented to it, and should implement | |
277 | <command>sysvinit</command> compatible signal handling (specifically: it needs to reboot on SIGINT, reexecute | |
278 | on SIGTERM, reload configuration on SIGHUP, and so on). With <option>--as-pid2</option> a minimal stub init | |
279 | process is run as PID 1 and the selected binary is executed as PID 2 (and hence does not need to implement any | |
280 | special semantics). The stub init process will reap processes as necessary and react appropriately to | |
281 | signals. It is recommended to use this mode to invoke arbitrary commands in containers, unless they have been | |
282 | modified to run correctly as PID 1. Or in other words: this switch should be used for pretty much all commands, | |
283 | except when the command refers to an init or shell implementation, as these are generally capable of running | |
284 | correctly as PID 1. This option may not be combined with <option>--boot</option>.</para> | |
285 | </listitem> | |
286 | </varlistentry> | |
287 | ||
288 | <varlistentry> | |
289 | <term><option>-b</option></term> | |
290 | <term><option>--boot</option></term> | |
291 | ||
292 | <listitem><para>Automatically search for an init binary and invoke it as PID 1, instead of a shell or a user | |
293 | supplied program. If this option is used, arguments specified on the command line are used as arguments for the | |
294 | init binary. This option may not be combined with <option>--as-pid2</option>.</para> | |
295 | ||
296 | <para>The following table explains the different modes of invocation and relationship to | |
297 | <option>--as-pid2</option> (see above):</para> | |
298 | ||
299 | <table> | |
300 | <title>Invocation Mode</title> | |
301 | <tgroup cols='2' align='left' colsep='1' rowsep='1'> | |
302 | <colspec colname="switch" /> | |
303 | <colspec colname="explanation" /> | |
304 | <thead> | |
305 | <row> | |
306 | <entry>Switch</entry> | |
307 | <entry>Explanation</entry> | |
308 | </row> | |
309 | </thead> | |
310 | <tbody> | |
311 | <row> | |
312 | <entry>Neither <option>--as-pid2</option> nor <option>--boot</option> specified</entry> | |
313 | <entry>The passed parameters are interpreted as the command line, which is executed as PID 1 in the container.</entry> | |
314 | </row> | |
315 | ||
316 | <row> | |
317 | <entry><option>--as-pid2</option> specified</entry> | |
318 | <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> | |
319 | </row> | |
320 | ||
321 | <row> | |
322 | <entry><option>--boot</option> specified</entry> | |
323 | <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> | |
324 | </row> | |
325 | ||
326 | </tbody> | |
327 | </tgroup> | |
328 | </table> | |
329 | ||
330 | <para>Note that <option>--boot</option> is the default mode of operation if the | |
331 | <filename>systemd-nspawn@.service</filename> template unit file is used.</para> | |
332 | </listitem> | |
333 | </varlistentry> | |
334 | ||
335 | <varlistentry> | |
336 | <term><option>--chdir=</option></term> | |
337 | ||
338 | <listitem><para>Change to the specified working directory before invoking the process in the container. Expects | |
339 | an absolute path in the container's file system namespace.</para></listitem> | |
340 | </varlistentry> | |
341 | ||
342 | <varlistentry> | |
343 | <term><option>--pivot-root=</option></term> | |
344 | ||
345 | <listitem><para>Pivot the specified directory to <filename>/</filename> inside the container, and either unmount the | |
346 | container's old root, or pivot it to another specified directory. Takes one of: a path argument — in which case the | |
347 | specified path will be pivoted to <filename>/</filename> and the old root will be unmounted; or a colon-separated pair | |
348 | of new root path and pivot destination for the old root. The new root path will be pivoted to <filename>/</filename>, | |
349 | and the old <filename>/</filename> will be pivoted to the other directory. Both paths must be absolute, and are resolved | |
350 | in the container's file system namespace.</para> | |
351 | ||
352 | <para>This is for containers which have several bootable directories in them; for example, several | |
353 | <ulink url="https://ostree.readthedocs.io/en/latest/">OSTree</ulink> deployments. It emulates the behavior of | |
354 | the boot loader and initial RAM disk which normally select which directory to mount as the root and start the | |
355 | container's PID 1 in.</para></listitem> | |
356 | </varlistentry> | |
357 | ||
358 | <varlistentry> | |
359 | <term><option>-u</option></term> | |
360 | <term><option>--user=</option></term> | |
361 | ||
362 | <listitem><para>After transitioning into the container, change | |
363 | to the specified user-defined in the container's user | |
364 | database. Like all other systemd-nspawn features, this is not | |
365 | a security feature and provides protection against accidental | |
366 | destructive operations only.</para></listitem> | |
367 | </varlistentry> | |
368 | ||
369 | <varlistentry> | |
370 | <term><option>-M</option></term> | |
371 | <term><option>--machine=</option></term> | |
372 | ||
373 | <listitem><para>Sets the machine name for this container. This | |
374 | name may be used to identify this container during its runtime | |
375 | (for example in tools like | |
376 | <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry> | |
377 | and similar), and is used to initialize the container's | |
378 | hostname (which the container can choose to override, | |
379 | however). If not specified, the last component of the root | |
380 | directory path of the container is used, possibly suffixed | |
381 | with a random identifier in case <option>--ephemeral</option> | |
382 | mode is selected. If the root directory selected is the host's | |
383 | root directory the host's hostname is used as default | |
384 | instead.</para></listitem> | |
385 | </varlistentry> | |
386 | ||
387 | <varlistentry> | |
388 | <term><option>--uuid=</option></term> | |
389 | ||
390 | <listitem><para>Set the specified UUID for the container. The | |
391 | init system will initialize | |
392 | <filename>/etc/machine-id</filename> from this if this file is | |
393 | not set yet. Note that this option takes effect only if | |
394 | <filename>/etc/machine-id</filename> in the container is | |
395 | unpopulated.</para></listitem> | |
396 | </varlistentry> | |
397 | ||
398 | <varlistentry> | |
399 | <term><option>--slice=</option></term> | |
400 | ||
401 | <listitem><para>Make the container part of the specified slice, instead of the default | |
402 | <filename>machine.slice</filename>. This applies only if the machine is run in its own scope unit, i.e. if | |
403 | <option>--keep-unit</option> isn't used.</para> | |
404 | </listitem> | |
405 | </varlistentry> | |
406 | ||
407 | <varlistentry> | |
408 | <term><option>--property=</option></term> | |
409 | ||
410 | <listitem><para>Set a unit property on the scope unit to register for the machine. This applies only if the | |
411 | machine is run in its own scope unit, i.e. if <option>--keep-unit</option> isn't used. Takes unit property | |
412 | assignments in the same format as <command>systemctl set-property</command>. This is useful to set memory | |
413 | limits and similar for container.</para> | |
414 | </listitem> | |
415 | </varlistentry> | |
416 | ||
417 | <varlistentry> | |
418 | <term><option>--private-users=</option></term> | |
419 | ||
420 | <listitem><para>Controls user namespacing. If enabled, the container will run with its own private set of UNIX | |
421 | user and group ids (UIDs and GIDs). This involves mapping the private UIDs/GIDs used in the container (starting | |
422 | with the container's root user 0 and up) to a range of UIDs/GIDs on the host that are not used for other | |
423 | purposes (usually in the range beyond the host's UID/GID 65536). The parameter may be specified as follows:</para> | |
424 | ||
425 | <orderedlist> | |
426 | <listitem><para>If one or two colon-separated numbers are specified, user namespacing is turned on. The first | |
427 | parameter specifies the first host UID/GID to assign to the container, the second parameter specifies the | |
428 | number of host UIDs/GIDs to assign to the container. If the second parameter is omitted, 65536 UIDs/GIDs are | |
429 | assigned.</para></listitem> | |
430 | ||
431 | <listitem><para>If the parameter is omitted, or true, user namespacing is turned on. The UID/GID range to | |
432 | use is determined automatically from the file ownership of the root directory of the container's directory | |
433 | tree. To use this option, make sure to prepare the directory tree in advance, and ensure that all files and | |
434 | directories in it are owned by UIDs/GIDs in the range you'd like to use. Also, make sure that used file ACLs | |
435 | exclusively reference UIDs/GIDs in the appropriate range. If this mode is used the number of UIDs/GIDs | |
436 | assigned to the container for use is 65536, and the UID/GID of the root directory must be a multiple of | |
437 | 65536.</para></listitem> | |
438 | ||
439 | <listitem><para>If the parameter is false, user namespacing is turned off. This is the default.</para> | |
440 | </listitem> | |
441 | ||
442 | <listitem><para>The special value <literal>pick</literal> turns on user namespacing. In this case the UID/GID | |
443 | range is automatically chosen. As first step, the file owner of the root directory of the container's | |
444 | directory tree is read, and it is checked that it is currently not used by the system otherwise (in | |
445 | particular, that no other container is using it). If this check is successful, the UID/GID range determined | |
446 | this way is used, similar to the behavior if "yes" is specified. If the check is not successful (and thus | |
447 | the UID/GID range indicated in the root directory's file owner is already used elsewhere) a new – currently | |
448 | unused – UID/GID range of 65536 UIDs/GIDs is randomly chosen between the host UID/GIDs of 524288 and | |
449 | 1878982656, always starting at a multiple of 65536. This setting implies | |
450 | <option>--private-users-chown</option> (see below), which has the effect that the files and directories in | |
451 | the container's directory tree will be owned by the appropriate users of the range picked. Using this option | |
452 | makes user namespace behavior fully automatic. Note that the first invocation of a previously unused | |
453 | container image might result in picking a new UID/GID range for it, and thus in the (possibly expensive) file | |
454 | ownership adjustment operation. However, subsequent invocations of the container will be cheap (unless of | |
455 | course the picked UID/GID range is assigned to a different use by then).</para></listitem> | |
456 | </orderedlist> | |
457 | ||
458 | <para>It is recommended to assign at least 65536 UIDs/GIDs to each container, so that the usable UID/GID range in the | |
459 | container covers 16 bit. For best security, do not assign overlapping UID/GID ranges to multiple containers. It is | |
460 | hence a good idea to use the upper 16 bit of the host 32-bit UIDs/GIDs as container identifier, while the lower 16 | |
461 | bit encode the container UID/GID used. This is in fact the behavior enforced by the | |
462 | <option>--private-users=pick</option> option.</para> | |
463 | ||
464 | <para>When user namespaces are used, the GID range assigned to each container is always chosen identical to the | |
465 | UID range.</para> | |
466 | ||
467 | <para>In most cases, using <option>--private-users=pick</option> is the recommended option as it enhances | |
468 | container security massively and operates fully automatically in most cases.</para> | |
469 | ||
470 | <para>Note that the picked UID/GID range is not written to <filename>/etc/passwd</filename> or | |
471 | <filename>/etc/group</filename>. In fact, the allocation of the range is not stored persistently anywhere, | |
472 | except in the file ownership of the files and directories of the container.</para> | |
473 | ||
474 | <para>Note that when user namespacing is used file ownership on disk reflects this, and all of the container's | |
475 | files and directories are owned by the container's effective user and group IDs. This means that copying files | |
476 | from and to the container image requires correction of the numeric UID/GID values, according to the UID/GID | |
477 | shift applied.</para></listitem> | |
478 | </varlistentry> | |
479 | ||
480 | <varlistentry> | |
481 | <term><option>--private-users-chown</option></term> | |
482 | ||
483 | <listitem><para>If specified, all files and directories in the container's directory tree will adjusted so that | |
484 | they are owned to the appropriate UIDs/GIDs selected for the container (see above). This operation is | |
485 | potentially expensive, as it involves descending and iterating through the full directory tree of the | |
486 | container. Besides actual file ownership, file ACLs are adjusted as well.</para> | |
487 | ||
488 | <para>This option is implied if <option>--private-users=pick</option> is used. This option has no effect if | |
489 | user namespacing is not used.</para></listitem> | |
490 | </varlistentry> | |
491 | ||
492 | <varlistentry> | |
493 | <term><option>-U</option></term> | |
494 | ||
495 | <listitem><para>If the kernel supports the user namespaces feature, equivalent to | |
496 | <option>--private-users=pick --private-users-chown</option>, otherwise equivalent to | |
497 | <option>--private-users=no</option>.</para> | |
498 | ||
499 | <para>Note that <option>-U</option> is the default if the | |
500 | <filename>systemd-nspawn@.service</filename> template unit file is used.</para> | |
501 | ||
502 | <para>Note: it is possible to undo the effect of <option>--private-users-chown</option> (or | |
503 | <option>-U</option>) on the file system by redoing the operation with the first UID of 0:</para> | |
504 | ||
505 | <programlisting>systemd-nspawn … --private-users=0 --private-users-chown</programlisting> | |
506 | </listitem> | |
507 | </varlistentry> | |
508 | ||
509 | <varlistentry> | |
510 | <term><option>--private-network</option></term> | |
511 | ||
512 | <listitem><para>Disconnect networking of the container from | |
513 | the host. This makes all network interfaces unavailable in the | |
514 | container, with the exception of the loopback device and those | |
515 | specified with <option>--network-interface=</option> and | |
516 | configured with <option>--network-veth</option>. If this | |
517 | option is specified, the CAP_NET_ADMIN capability will be | |
518 | added to the set of capabilities the container retains. The | |
519 | latter may be disabled by using | |
520 | <option>--drop-capability=</option>.</para></listitem> | |
521 | </varlistentry> | |
522 | ||
523 | <varlistentry> | |
524 | <term><option>--network-interface=</option></term> | |
525 | ||
526 | <listitem><para>Assign the specified network interface to the | |
527 | container. This will remove the specified interface from the | |
528 | calling namespace and place it in the container. When the | |
529 | container terminates, it is moved back to the host namespace. | |
530 | Note that <option>--network-interface=</option> implies | |
531 | <option>--private-network</option>. This option may be used | |
532 | more than once to add multiple network interfaces to the | |
533 | container.</para></listitem> | |
534 | </varlistentry> | |
535 | ||
536 | <varlistentry> | |
537 | <term><option>--network-macvlan=</option></term> | |
538 | ||
539 | <listitem><para>Create a <literal>macvlan</literal> interface | |
540 | of the specified Ethernet network interface and add it to the | |
541 | container. A <literal>macvlan</literal> interface is a virtual | |
542 | interface that adds a second MAC address to an existing | |
543 | physical Ethernet link. The interface in the container will be | |
544 | named after the interface on the host, prefixed with | |
545 | <literal>mv-</literal>. Note that | |
546 | <option>--network-macvlan=</option> implies | |
547 | <option>--private-network</option>. This option may be used | |
548 | more than once to add multiple network interfaces to the | |
549 | container.</para></listitem> | |
550 | </varlistentry> | |
551 | ||
552 | <varlistentry> | |
553 | <term><option>--network-ipvlan=</option></term> | |
554 | ||
555 | <listitem><para>Create an <literal>ipvlan</literal> interface | |
556 | of the specified Ethernet network interface and add it to the | |
557 | container. An <literal>ipvlan</literal> interface is a virtual | |
558 | interface, similar to a <literal>macvlan</literal> interface, | |
559 | which uses the same MAC address as the underlying interface. | |
560 | The interface in the container will be named after the | |
561 | interface on the host, prefixed with <literal>iv-</literal>. | |
562 | Note that <option>--network-ipvlan=</option> implies | |
563 | <option>--private-network</option>. This option may be used | |
564 | more than once to add multiple network interfaces to the | |
565 | container.</para></listitem> | |
566 | </varlistentry> | |
567 | ||
568 | <varlistentry> | |
569 | <term><option>-n</option></term> | |
570 | <term><option>--network-veth</option></term> | |
571 | ||
572 | <listitem><para>Create a virtual Ethernet link (<literal>veth</literal>) between host and container. The host | |
573 | side of the Ethernet link will be available as a network interface named after the container's name (as | |
574 | specified with <option>--machine=</option>), prefixed with <literal>ve-</literal>. The container side of the | |
575 | Ethernet link will be named <literal>host0</literal>. The <option>--network-veth</option> option implies | |
576 | <option>--private-network</option>.</para> | |
577 | ||
578 | <para>Note that | |
579 | <citerefentry><refentrytitle>systemd-networkd.service</refentrytitle><manvolnum>8</manvolnum></citerefentry> | |
580 | includes by default a network file <filename>/usr/lib/systemd/network/80-container-ve.network</filename> | |
581 | matching the host-side interfaces created this way, which contains settings to enable automatic address | |
582 | provisioning on the created virtual link via DHCP, as well as automatic IP routing onto the host's external | |
583 | network interfaces. It also contains <filename>/usr/lib/systemd/network/80-container-host0.network</filename> | |
584 | matching the container-side interface created this way, containing settings to enable client side address | |
585 | assignment via DHCP. In case <filename>systemd-networkd</filename> is running on both the host and inside the | |
586 | container, automatic IP communication from the container to the host is thus available, with further | |
587 | connectivity to the external network.</para> | |
588 | ||
589 | <para>Note that <option>--network-veth</option> is the default if the | |
590 | <filename>systemd-nspawn@.service</filename> template unit file is used.</para> | |
591 | </listitem> | |
592 | </varlistentry> | |
593 | ||
594 | <varlistentry> | |
595 | <term><option>--network-veth-extra=</option></term> | |
596 | ||
597 | <listitem><para>Adds an additional virtual Ethernet link | |
598 | between host and container. Takes a colon-separated pair of | |
599 | host interface name and container interface name. The latter | |
600 | may be omitted in which case the container and host sides will | |
601 | be assigned the same name. This switch is independent of | |
602 | <option>--network-veth</option>, and — in contrast — may be | |
603 | used multiple times, and allows configuration of the network | |
604 | interface names. Note that <option>--network-bridge=</option> | |
605 | has no effect on interfaces created with | |
606 | <option>--network-veth-extra=</option>.</para></listitem> | |
607 | </varlistentry> | |
608 | ||
609 | <varlistentry> | |
610 | <term><option>--network-bridge=</option></term> | |
611 | ||
612 | <listitem><para>Adds the host side of the Ethernet link created with <option>--network-veth</option> to the | |
613 | specified Ethernet bridge interface. Expects a valid network interface name of a bridge device as | |
614 | argument. Note that <option>--network-bridge=</option> implies <option>--network-veth</option>. If this option | |
615 | is used, the host side of the Ethernet link will use the <literal>vb-</literal> prefix instead of | |
616 | <literal>ve-</literal>.</para></listitem> | |
617 | </varlistentry> | |
618 | ||
619 | <varlistentry> | |
620 | <term><option>--network-zone=</option></term> | |
621 | ||
622 | <listitem><para>Creates a virtual Ethernet link (<literal>veth</literal>) to the container and adds it to an | |
623 | automatically managed Ethernet bridge interface. The bridge interface is named after the passed argument, | |
624 | prefixed with <literal>vz-</literal>. The bridge interface is automatically created when the first container | |
625 | configured for its name is started, and is automatically removed when the last container configured for its | |
626 | name exits. Hence, each bridge interface configured this way exists only as long as there's at least one | |
627 | container referencing it running. This option is very similar to <option>--network-bridge=</option>, besides | |
628 | this automatic creation/removal of the bridge device.</para> | |
629 | ||
630 | <para>This setting makes it easy to place multiple related containers on a common, virtual Ethernet-based | |
631 | broadcast domain, here called a "zone". Each container may only be part of one zone, but each zone may contain | |
632 | any number of containers. Each zone is referenced by its name. Names may be chosen freely (as long as they form | |
633 | valid network interface names when prefixed with <literal>vz-</literal>), and it is sufficient to pass the same | |
634 | name to the <option>--network-zone=</option> switch of the various concurrently running containers to join | |
635 | them in one zone.</para> | |
636 | ||
637 | <para>Note that | |
638 | <citerefentry><refentrytitle>systemd-networkd.service</refentrytitle><manvolnum>8</manvolnum></citerefentry> | |
639 | includes by default a network file <filename>/usr/lib/systemd/network/80-container-vz.network</filename> | |
640 | matching the bridge interfaces created this way, which contains settings to enable automatic address | |
641 | provisioning on the created virtual network via DHCP, as well as automatic IP routing onto the host's external | |
642 | network interfaces. Using <option>--network-zone=</option> is hence in most cases fully automatic and | |
643 | sufficient to connect multiple local containers in a joined broadcast domain to the host, with further | |
644 | connectivity to the external network.</para> | |
645 | </listitem> | |
646 | </varlistentry> | |
647 | ||
648 | <varlistentry> | |
649 | <term><option>-p</option></term> | |
650 | <term><option>--port=</option></term> | |
651 | ||
652 | <listitem><para>If private networking is enabled, maps an IP | |
653 | port on the host onto an IP port on the container. Takes a | |
654 | protocol specifier (either <literal>tcp</literal> or | |
655 | <literal>udp</literal>), separated by a colon from a host port | |
656 | number in the range 1 to 65535, separated by a colon from a | |
657 | container port number in the range from 1 to 65535. The | |
658 | protocol specifier and its separating colon may be omitted, in | |
659 | which case <literal>tcp</literal> is assumed. The container | |
660 | port number and its colon may be omitted, in which case the | |
661 | same port as the host port is implied. This option is only | |
662 | supported if private networking is used, such as with | |
663 | <option>--network-veth</option>, <option>--network-zone=</option> | |
664 | <option>--network-bridge=</option>.</para></listitem> | |
665 | </varlistentry> | |
666 | ||
667 | <varlistentry> | |
668 | <term><option>-Z</option></term> | |
669 | <term><option>--selinux-context=</option></term> | |
670 | ||
671 | <listitem><para>Sets the SELinux security context to be used | |
672 | to label processes in the container.</para> | |
673 | </listitem> | |
674 | </varlistentry> | |
675 | ||
676 | <varlistentry> | |
677 | <term><option>-L</option></term> | |
678 | <term><option>--selinux-apifs-context=</option></term> | |
679 | ||
680 | <listitem><para>Sets the SELinux security context to be used | |
681 | to label files in the virtual API file systems in the | |
682 | container.</para> | |
683 | </listitem> | |
684 | </varlistentry> | |
685 | ||
686 | <varlistentry> | |
687 | <term><option>--capability=</option></term> | |
688 | ||
689 | <listitem><para>List one or more additional capabilities to | |
690 | grant the container. Takes a comma-separated list of | |
691 | capability names, see | |
692 | <citerefentry project='man-pages'><refentrytitle>capabilities</refentrytitle><manvolnum>7</manvolnum></citerefentry> | |
693 | for more information. Note that the following capabilities | |
694 | will be granted in any way: CAP_CHOWN, CAP_DAC_OVERRIDE, | |
695 | CAP_DAC_READ_SEARCH, CAP_FOWNER, CAP_FSETID, CAP_IPC_OWNER, | |
696 | CAP_KILL, CAP_LEASE, CAP_LINUX_IMMUTABLE, | |
697 | CAP_NET_BIND_SERVICE, CAP_NET_BROADCAST, CAP_NET_RAW, | |
698 | CAP_SETGID, CAP_SETFCAP, CAP_SETPCAP, CAP_SETUID, | |
699 | CAP_SYS_ADMIN, CAP_SYS_CHROOT, CAP_SYS_NICE, CAP_SYS_PTRACE, | |
700 | CAP_SYS_TTY_CONFIG, CAP_SYS_RESOURCE, CAP_SYS_BOOT, | |
701 | CAP_AUDIT_WRITE, CAP_AUDIT_CONTROL. Also CAP_NET_ADMIN is | |
702 | retained if <option>--private-network</option> is specified. | |
703 | If the special value <literal>all</literal> is passed, all | |
704 | capabilities are retained.</para></listitem> | |
705 | </varlistentry> | |
706 | ||
707 | <varlistentry> | |
708 | <term><option>--drop-capability=</option></term> | |
709 | ||
710 | <listitem><para>Specify one or more additional capabilities to | |
711 | drop for the container. This allows running the container with | |
712 | fewer capabilities than the default (see | |
713 | above).</para></listitem> | |
714 | </varlistentry> | |
715 | ||
716 | <varlistentry> | |
717 | <term><option>--system-call-filter=</option></term> | |
718 | ||
719 | <listitem><para>Alter the system call filter applied to containers. Takes a space-separated list of system call | |
720 | names or group names (the latter prefixed with <literal>@</literal>, as listed by the | |
721 | <command>syscall-filter</command> command of <citerefentry | |
722 | project='man-pages'><refentrytitle>systemd-analyze</refentrytitle><manvolnum>1</manvolnum></citerefentry>). Passed | |
723 | system calls will be permitted. The list may optionally be prefixed by <literal>~</literal>, in which case all | |
724 | listed system calls are prohibited. If this command line option is used multiple times the configured lists are | |
725 | combined. If both a positive and a negative list (that is one system call list without and one with the | |
726 | <literal>~</literal> prefix) are configured, the negative list takes precedence over the positive list. Note | |
727 | that <command>systemd-nspawn</command> always implements a system call whitelist (as opposed to a blacklist), | |
728 | and this command line option hence adds or removes entries from the default whitelist, depending on the | |
729 | <literal>~</literal> prefix. Note that the applied system call filter is also altered implicitly if additional | |
730 | capabilities are passed using the <command>--capabilities=</command>.</para></listitem> | |
731 | </varlistentry> | |
732 | ||
733 | <varlistentry> | |
734 | <term><option>--kill-signal=</option></term> | |
735 | ||
736 | <listitem><para>Specify the process signal to send to the | |
737 | container's PID 1 when nspawn itself receives SIGTERM, in | |
738 | order to trigger an orderly shutdown of the | |
739 | container. Defaults to SIGRTMIN+3 if <option>--boot</option> | |
740 | is used (on systemd-compatible init systems SIGRTMIN+3 | |
741 | triggers an orderly shutdown). For a list of valid signals, see | |
742 | <citerefentry project='man-pages'><refentrytitle>signal</refentrytitle><manvolnum>7</manvolnum></citerefentry>.</para></listitem> | |
743 | </varlistentry> | |
744 | ||
745 | <varlistentry> | |
746 | <term><option>--link-journal=</option></term> | |
747 | ||
748 | <listitem><para>Control whether the container's journal shall | |
749 | be made visible to the host system. If enabled, allows viewing | |
750 | the container's journal files from the host (but not vice | |
751 | versa). Takes one of <literal>no</literal>, | |
752 | <literal>host</literal>, <literal>try-host</literal>, | |
753 | <literal>guest</literal>, <literal>try-guest</literal>, | |
754 | <literal>auto</literal>. If <literal>no</literal>, the journal | |
755 | is not linked. If <literal>host</literal>, the journal files | |
756 | are stored on the host file system (beneath | |
757 | <filename>/var/log/journal/<replaceable>machine-id</replaceable></filename>) | |
758 | and the subdirectory is bind-mounted into the container at the | |
759 | same location. If <literal>guest</literal>, the journal files | |
760 | are stored on the guest file system (beneath | |
761 | <filename>/var/log/journal/<replaceable>machine-id</replaceable></filename>) | |
762 | and the subdirectory is symlinked into the host at the same | |
763 | location. <literal>try-host</literal> and | |
764 | <literal>try-guest</literal> do the same but do not fail if | |
765 | the host does not have persistent journaling enabled. If | |
766 | <literal>auto</literal> (the default), and the right | |
767 | subdirectory of <filename>/var/log/journal</filename> exists, | |
768 | it will be bind mounted into the container. If the | |
769 | subdirectory does not exist, no linking is performed. | |
770 | Effectively, booting a container once with | |
771 | <literal>guest</literal> or <literal>host</literal> will link | |
772 | the journal persistently if further on the default of | |
773 | <literal>auto</literal> is used.</para> | |
774 | ||
775 | <para>Note that <option>--link-journal=try-guest</option> is the default if the | |
776 | <filename>systemd-nspawn@.service</filename> template unit file is used.</para></listitem> | |
777 | </varlistentry> | |
778 | ||
779 | <varlistentry> | |
780 | <term><option>-j</option></term> | |
781 | ||
782 | <listitem><para>Equivalent to | |
783 | <option>--link-journal=try-guest</option>.</para></listitem> | |
784 | </varlistentry> | |
785 | ||
786 | <varlistentry> | |
787 | <term><option>--read-only</option></term> | |
788 | ||
789 | <listitem><para>Mount the root file system read-only for the | |
790 | container.</para></listitem> | |
791 | </varlistentry> | |
792 | ||
793 | <varlistentry> | |
794 | <term><option>--bind=</option></term> | |
795 | <term><option>--bind-ro=</option></term> | |
796 | ||
797 | <listitem><para>Bind mount a file or directory from the host into the container. Takes one of: a path | |
798 | argument — in which case the specified path will be mounted from the host to the same path in the container, or | |
799 | a colon-separated pair of paths — in which case the first specified path is the source in the host, and the | |
800 | second path is the destination in the container, or a colon-separated triple of source path, destination path | |
801 | and mount options. The source path may optionally be prefixed with a <literal>+</literal> character. If so, the | |
802 | source path is taken relative to the image's root directory. This permits setting up bind mounts within the | |
803 | container image. The source path may be specified as empty string, in which case a temporary directory below | |
804 | the host's <filename>/var/tmp</filename> directory is used. It is automatically removed when the container is | |
805 | shut down. Mount options are comma-separated and currently, only <option>rbind</option> and | |
806 | <option>norbind</option> are allowed, controlling whether to create a recursive or a regular bind | |
807 | mount. Defaults to "rbind". Backslash escapes are interpreted, so <literal>\:</literal> may be used to embed | |
808 | colons in either path. This option may be specified multiple times for creating multiple independent bind | |
809 | mount points. The <option>--bind-ro=</option> option creates read-only bind mounts.</para></listitem> | |
810 | </varlistentry> | |
811 | ||
812 | <varlistentry> | |
813 | <term><option>--tmpfs=</option></term> | |
814 | ||
815 | <listitem><para>Mount a tmpfs file system into the container. | |
816 | Takes a single absolute path argument that specifies where to | |
817 | mount the tmpfs instance to (in which case the directory | |
818 | access mode will be chosen as 0755, owned by root/root), or | |
819 | optionally a colon-separated pair of path and mount option | |
820 | string that is used for mounting (in which case the kernel | |
821 | default for access mode and owner will be chosen, unless | |
822 | otherwise specified). This option is particularly useful for | |
823 | mounting directories such as <filename>/var</filename> as | |
824 | tmpfs, to allow state-less systems, in particular when | |
825 | combined with <option>--read-only</option>. | |
826 | Backslash escapes are interpreted in the path, so | |
827 | <literal>\:</literal> may be used to embed colons in the path. | |
828 | </para></listitem> | |
829 | </varlistentry> | |
830 | ||
831 | <varlistentry> | |
832 | <term><option>--overlay=</option></term> | |
833 | <term><option>--overlay-ro=</option></term> | |
834 | ||
835 | <listitem><para>Combine multiple directory trees into one | |
836 | overlay file system and mount it into the container. Takes a | |
837 | list of colon-separated paths to the directory trees to | |
838 | combine and the destination mount point.</para> | |
839 | ||
840 | <para>Backslash escapes are interpreted in the paths, so | |
841 | <literal>\:</literal> may be used to embed colons in the paths. | |
842 | </para> | |
843 | ||
844 | <para>If three or more paths are specified, then the last | |
845 | specified path is the destination mount point in the | |
846 | container, all paths specified before refer to directory trees | |
847 | on the host and are combined in the specified order into one | |
848 | overlay file system. The left-most path is hence the lowest | |
849 | directory tree, the second-to-last path the highest directory | |
850 | tree in the stacking order. If <option>--overlay-ro=</option> | |
851 | is used instead of <option>--overlay=</option>, a read-only | |
852 | overlay file system is created. If a writable overlay file | |
853 | system is created, all changes made to it are written to the | |
854 | highest directory tree in the stacking order, i.e. the | |
855 | second-to-last specified.</para> | |
856 | ||
857 | <para>If only two paths are specified, then the second | |
858 | specified path is used both as the top-level directory tree in | |
859 | the stacking order as seen from the host, as well as the mount | |
860 | point for the overlay file system in the container. At least | |
861 | two paths have to be specified.</para> | |
862 | ||
863 | <para>The source paths may optionally be prefixed with <literal>+</literal> character. If so they are taken | |
864 | relative to the image's root directory. The uppermost source path may also be specified as empty string, in | |
865 | which case a temporary directory below the host's <filename>/var/tmp</filename> is used. The directory is | |
866 | removed automatically when the container is shut down. This behaviour is useful in order to make read-only | |
867 | container directories writable while the container is running. For example, use the | |
868 | <literal>--overlay=+/var::/var</literal> option in order to automatically overlay a writable temporary | |
869 | directory on a read-only <filename>/var</filename> directory.</para> | |
870 | ||
871 | <para>For details about overlay file systems, see <ulink | |
872 | url="https://www.kernel.org/doc/Documentation/filesystems/overlayfs.txt">overlayfs.txt</ulink>. Note | |
873 | that the semantics of overlay file systems are substantially | |
874 | different from normal file systems, in particular regarding | |
875 | reported device and inode information. Device and inode | |
876 | information may change for a file while it is being written | |
877 | to, and processes might see out-of-date versions of files at | |
878 | times. Note that this switch automatically derives the | |
879 | <literal>workdir=</literal> mount option for the overlay file | |
880 | system from the top-level directory tree, making it a sibling | |
881 | of it. It is hence essential that the top-level directory tree | |
882 | is not a mount point itself (since the working directory must | |
883 | be on the same file system as the top-most directory | |
884 | tree). Also note that the <literal>lowerdir=</literal> mount | |
885 | option receives the paths to stack in the opposite order of | |
886 | this switch.</para></listitem> | |
887 | </varlistentry> | |
888 | ||
889 | <varlistentry> | |
890 | <term><option>-E <replaceable>NAME</replaceable>=<replaceable>VALUE</replaceable></option></term> | |
891 | <term><option>--setenv=<replaceable>NAME</replaceable>=<replaceable>VALUE</replaceable></option></term> | |
892 | ||
893 | <listitem><para>Specifies an environment variable assignment | |
894 | to pass to the init process in the container, in the format | |
895 | <literal>NAME=VALUE</literal>. This may be used to override | |
896 | the default variables or to set additional variables. This | |
897 | parameter may be used more than once.</para></listitem> | |
898 | </varlistentry> | |
899 | ||
900 | <varlistentry> | |
901 | <term><option>--register=</option></term> | |
902 | ||
903 | <listitem><para>Controls whether the container is registered with | |
904 | <citerefentry><refentrytitle>systemd-machined</refentrytitle><manvolnum>8</manvolnum></citerefentry>. Takes a | |
905 | boolean argument, which defaults to <literal>yes</literal>. This option should be enabled when the container | |
906 | runs a full Operating System (more specifically: a system and service manager as PID 1), and is useful to | |
907 | ensure that the container is accessible via | |
908 | <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry> and shown by | |
909 | tools such as <citerefentry | |
910 | project='man-pages'><refentrytitle>ps</refentrytitle><manvolnum>1</manvolnum></citerefentry>. If the container | |
911 | does not run a service manager, it is recommended to set this option to | |
912 | <literal>no</literal>.</para></listitem> | |
913 | </varlistentry> | |
914 | ||
915 | <varlistentry> | |
916 | <term><option>--keep-unit</option></term> | |
917 | ||
918 | <listitem><para>Instead of creating a transient scope unit to run the container in, simply use the service or | |
919 | scope unit <command>systemd-nspawn</command> has been invoked in. If <option>--register=yes</option> is set | |
920 | this unit is registered with | |
921 | <citerefentry><refentrytitle>systemd-machined</refentrytitle><manvolnum>8</manvolnum></citerefentry>. This | |
922 | switch should be used if <command>systemd-nspawn</command> is invoked from within a service unit, and the | |
923 | service unit's sole purpose is to run a single <command>systemd-nspawn</command> container. This option is not | |
924 | available if run from a user session.</para> | |
925 | <para>Note that passing <option>--keep-unit</option> disables the effect of <option>--slice=</option> and | |
926 | <option>--property=</option>. Use <option>--keep-unit</option> and <option>--register=no</option> in | |
927 | combination to disable any kind of unit allocation or registration with | |
928 | <command>systemd-machined</command>.</para></listitem> | |
929 | </varlistentry> | |
930 | ||
931 | <varlistentry> | |
932 | <term><option>--personality=</option></term> | |
933 | ||
934 | <listitem><para>Control the architecture ("personality") | |
935 | reported by | |
936 | <citerefentry project='man-pages'><refentrytitle>uname</refentrytitle><manvolnum>2</manvolnum></citerefentry> | |
937 | in the container. Currently, only <literal>x86</literal> and | |
938 | <literal>x86-64</literal> are supported. This is useful when | |
939 | running a 32-bit container on a 64-bit host. If this setting | |
940 | is not used, the personality reported in the container is the | |
941 | same as the one reported on the host.</para></listitem> | |
942 | </varlistentry> | |
943 | ||
944 | <varlistentry> | |
945 | <term><option>-q</option></term> | |
946 | <term><option>--quiet</option></term> | |
947 | ||
948 | <listitem><para>Turns off any status output by the tool | |
949 | itself. When this switch is used, the only output from nspawn | |
950 | will be the console output of the container OS | |
951 | itself.</para></listitem> | |
952 | </varlistentry> | |
953 | ||
954 | <varlistentry> | |
955 | <term><option>--volatile</option></term> | |
956 | <term><option>--volatile=</option><replaceable>MODE</replaceable></term> | |
957 | ||
958 | <listitem><para>Boots the container in volatile mode. When no | |
959 | mode parameter is passed or when mode is specified as | |
960 | <option>yes</option>, full volatile mode is enabled. This | |
961 | means the root directory is mounted as a mostly unpopulated | |
962 | <literal>tmpfs</literal> instance, and | |
963 | <filename>/usr</filename> from the OS tree is mounted into it | |
964 | in read-only mode (the system thus starts up with read-only OS | |
965 | image, but pristine state and configuration, any changes | |
966 | are lost on shutdown). When the mode parameter | |
967 | is specified as <option>state</option>, the OS tree is | |
968 | mounted read-only, but <filename>/var</filename> is mounted as | |
969 | a <literal>tmpfs</literal> instance into it (the system thus | |
970 | starts up with read-only OS resources and configuration, but | |
971 | pristine state, and any changes to the latter are lost on | |
972 | shutdown). When the mode parameter is specified as | |
973 | <option>no</option> (the default), the whole OS tree is made | |
974 | available writable.</para> | |
975 | ||
976 | <para>This option provides similar functionality for containers as the <literal>systemd.volatile=</literal> | |
977 | kernel command line switch provides for host systems. See | |
978 | <citerefentry><refentrytitle>kernel-command-line</refentrytitle><manvolnum>7</manvolnum></citerefentry> for | |
979 | details.</para> | |
980 | ||
981 | <para>Note that enabling this setting will only work correctly with operating systems in the container that can | |
982 | boot up with only <filename>/usr</filename> mounted, and are able to automatically populate | |
983 | <filename>/var</filename>, and also <filename>/etc</filename> in case of | |
984 | <literal>--volatile=yes</literal>.</para></listitem> | |
985 | </varlistentry> | |
986 | ||
987 | <varlistentry> | |
988 | <term><option>--settings=</option><replaceable>MODE</replaceable></term> | |
989 | ||
990 | <listitem><para>Controls whether | |
991 | <command>systemd-nspawn</command> shall search for and use | |
992 | additional per-container settings from | |
993 | <filename>.nspawn</filename> files. Takes a boolean or the | |
994 | special values <option>override</option> or | |
995 | <option>trusted</option>.</para> | |
996 | ||
997 | <para>If enabled (the default), a settings file named after the | |
998 | machine (as specified with the <option>--machine=</option> | |
999 | setting, or derived from the directory or image file name) | |
1000 | with the suffix <filename>.nspawn</filename> is searched in | |
1001 | <filename>/etc/systemd/nspawn/</filename> and | |
1002 | <filename>/run/systemd/nspawn/</filename>. If it is found | |
1003 | there, its settings are read and used. If it is not found | |
1004 | there, it is subsequently searched in the same directory as the | |
1005 | image file or in the immediate parent of the root directory of | |
1006 | the container. In this case, if the file is found, its settings | |
1007 | will be also read and used, but potentially unsafe settings | |
1008 | are ignored. Note that in both these cases, settings on the | |
1009 | command line take precedence over the corresponding settings | |
1010 | from loaded <filename>.nspawn</filename> files, if both are | |
1011 | specified. Unsafe settings are considered all settings that | |
1012 | elevate the container's privileges or grant access to | |
1013 | additional resources such as files or directories of the | |
1014 | host. For details about the format and contents of | |
1015 | <filename>.nspawn</filename> files, consult | |
1016 | <citerefentry><refentrytitle>systemd.nspawn</refentrytitle><manvolnum>5</manvolnum></citerefentry>.</para> | |
1017 | ||
1018 | <para>If this option is set to <option>override</option>, the | |
1019 | file is searched, read and used the same way, however, the order of | |
1020 | precedence is reversed: settings read from the | |
1021 | <filename>.nspawn</filename> file will take precedence over | |
1022 | the corresponding command line options, if both are | |
1023 | specified.</para> | |
1024 | ||
1025 | <para>If this option is set to <option>trusted</option>, the | |
1026 | file is searched, read and used the same way, but regardless | |
1027 | of being found in <filename>/etc/systemd/nspawn/</filename>, | |
1028 | <filename>/run/systemd/nspawn/</filename> or next to the image | |
1029 | file or container root directory, all settings will take | |
1030 | effect, however, command line arguments still take precedence | |
1031 | over corresponding settings.</para> | |
1032 | ||
1033 | <para>If disabled, no <filename>.nspawn</filename> file is read | |
1034 | and no settings except the ones on the command line are in | |
1035 | effect.</para></listitem> | |
1036 | </varlistentry> | |
1037 | ||
1038 | <varlistentry> | |
1039 | <term><option>--notify-ready=</option></term> | |
1040 | ||
1041 | <listitem><para>Configures support for notifications from the container's init process. | |
1042 | <option>--notify-ready=</option> takes a boolean (<option>no</option> and <option>yes</option>). | |
1043 | With option <option>no</option> systemd-nspawn notifies systemd | |
1044 | with a <literal>READY=1</literal> message when the init process is created. | |
1045 | With option <option>yes</option> systemd-nspawn waits for the | |
1046 | <literal>READY=1</literal> message from the init process in the container | |
1047 | before sending its own to systemd. For more details about notifications | |
1048 | see <citerefentry><refentrytitle>sd_notify</refentrytitle><manvolnum>3</manvolnum></citerefentry>).</para></listitem> | |
1049 | </varlistentry> | |
1050 | ||
1051 | <xi:include href="standard-options.xml" xpointer="help" /> | |
1052 | <xi:include href="standard-options.xml" xpointer="version" /> | |
1053 | </variablelist> | |
1054 | ||
1055 | </refsect1> | |
1056 | ||
1057 | <refsect1> | |
1058 | <title>Examples</title> | |
1059 | ||
1060 | <example> | |
1061 | <title>Download a | |
1062 | <ulink url="https://getfedora.org">Fedora</ulink> image and start a shell in it</title> | |
1063 | ||
1064 | <programlisting># machinectl pull-raw --verify=no \ | |
1065 | https://download.fedoraproject.org/pub/fedora/linux/releases/25/CloudImages/x86_64/images/Fedora-Cloud-Base-25-1.3.x86_64.raw.xz | |
1066 | # systemd-nspawn -M Fedora-Cloud-Base-25-1.3.x86_64.raw</programlisting> | |
1067 | ||
1068 | <para>This downloads an image using | |
1069 | <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry> | |
1070 | and opens a shell in it.</para> | |
1071 | </example> | |
1072 | ||
1073 | <example> | |
1074 | <title>Build and boot a minimal Fedora distribution in a container</title> | |
1075 | ||
1076 | <programlisting># dnf -y --releasever=27 --installroot=/var/lib/machines/f27container \ | |
1077 | --disablerepo='*' --enablerepo=fedora --enablerepo=updates install \ | |
1078 | systemd passwd dnf fedora-release vim-minimal | |
1079 | # systemd-nspawn -bD /var/lib/machines/f27container</programlisting> | |
1080 | ||
1081 | <para>This installs a minimal Fedora distribution into the | |
1082 | directory <filename noindex='true'>/var/lib/machines/f27container</filename> | |
1083 | and then boots an OS in a namespace container in it. Because the installation | |
1084 | is located underneath the standard <filename>/var/lib/machines/</filename> | |
1085 | directory, it is also possible to start the machine using | |
1086 | <command>systemd-nspawn -M f27container</command>.</para> | |
1087 | </example> | |
1088 | ||
1089 | <example> | |
1090 | <title>Spawn a shell in a container of a minimal Debian unstable distribution</title> | |
1091 | ||
1092 | <programlisting># debootstrap --arch=amd64 unstable ~/debian-tree/ | |
1093 | # systemd-nspawn -D ~/debian-tree/</programlisting> | |
1094 | ||
1095 | <para>This installs a minimal Debian unstable distribution into | |
1096 | the directory <filename>~/debian-tree/</filename> and then | |
1097 | spawns a shell in a namespace container in it.</para> | |
1098 | ||
1099 | <para><command>debootstrap</command> supports | |
1100 | <ulink url="https://www.debian.org">Debian</ulink>, | |
1101 | <ulink url="https://www.ubuntu.com">Ubuntu</ulink>, | |
1102 | and <ulink url="https://www.tanglu.org">Tanglu</ulink> | |
1103 | out of the box, so the same command can be used to install any of those. For other | |
1104 | distributions from the Debian family, a mirror has to be specified, see | |
1105 | <citerefentry project='die-net'><refentrytitle>debootstrap</refentrytitle><manvolnum>8</manvolnum></citerefentry>. | |
1106 | </para> | |
1107 | </example> | |
1108 | ||
1109 | <example> | |
1110 | <title>Boot a minimal | |
1111 | <ulink url="https://www.archlinux.org">Arch Linux</ulink> distribution in a container</title> | |
1112 | ||
1113 | <programlisting># pacstrap -c -d ~/arch-tree/ base | |
1114 | # systemd-nspawn -bD ~/arch-tree/</programlisting> | |
1115 | ||
1116 | <para>This installs a minimal Arch Linux distribution into the | |
1117 | directory <filename>~/arch-tree/</filename> and then boots an OS | |
1118 | in a namespace container in it.</para> | |
1119 | </example> | |
1120 | ||
1121 | <example> | |
1122 | <title>Install the | |
1123 | <ulink url="https://software.opensuse.org/distributions/tumbleweed">OpenSUSE Tumbleweed</ulink> | |
1124 | rolling distribution</title> | |
1125 | ||
1126 | <programlisting># zypper --root=/var/lib/machines/tumbleweed ar -c \ | |
1127 | https://download.opensuse.org/tumbleweed/repo/oss tumbleweed | |
1128 | # zypper --root=/var/lib/machines/tumbleweed refresh | |
1129 | # zypper --root=/var/lib/machines/tumbleweed install --no-recommends \ | |
1130 | systemd shadow zypper openSUSE-release vim | |
1131 | # systemd-nspawn -M tumbleweed passwd root | |
1132 | # systemd-nspawn -M tumbleweed -b</programlisting> | |
1133 | </example> | |
1134 | ||
1135 | <example> | |
1136 | <title>Boot into an ephemeral snapshot of the host system</title> | |
1137 | ||
1138 | <programlisting># systemd-nspawn -D / -xb</programlisting> | |
1139 | ||
1140 | <para>This runs a copy of the host system in a snapshot which is removed immediately when the container | |
1141 | exits. All file system changes made during runtime will be lost on shutdown, hence.</para> | |
1142 | </example> | |
1143 | ||
1144 | <example> | |
1145 | <title>Run a container with SELinux sandbox security contexts</title> | |
1146 | ||
1147 | <programlisting># chcon system_u:object_r:svirt_sandbox_file_t:s0:c0,c1 -R /srv/container | |
1148 | # systemd-nspawn -L system_u:object_r:svirt_sandbox_file_t:s0:c0,c1 \ | |
1149 | -Z system_u:system_r:svirt_lxc_net_t:s0:c0,c1 -D /srv/container /bin/sh</programlisting> | |
1150 | </example> | |
1151 | ||
1152 | <example> | |
1153 | <title>Run a container with an OSTree deployment</title> | |
1154 | ||
1155 | <programlisting># systemd-nspawn -b -i ~/image.raw \ | |
1156 | --pivot-root=/ostree/deploy/$OS/deploy/$CHECKSUM:/sysroot \ | |
1157 | --bind=+/sysroot/ostree/deploy/$OS/var:/var</programlisting> | |
1158 | </example> | |
1159 | </refsect1> | |
1160 | ||
1161 | <refsect1> | |
1162 | <title>Exit status</title> | |
1163 | ||
1164 | <para>The exit code of the program executed in the container is | |
1165 | returned.</para> | |
1166 | </refsect1> | |
1167 | ||
1168 | <refsect1> | |
1169 | <title>See Also</title> | |
1170 | <para> | |
1171 | <citerefentry><refentrytitle>systemd</refentrytitle><manvolnum>1</manvolnum></citerefentry>, | |
1172 | <citerefentry><refentrytitle>systemd.nspawn</refentrytitle><manvolnum>5</manvolnum></citerefentry>, | |
1173 | <citerefentry project='man-pages'><refentrytitle>chroot</refentrytitle><manvolnum>1</manvolnum></citerefentry>, | |
1174 | <citerefentry project='mankier'><refentrytitle>dnf</refentrytitle><manvolnum>8</manvolnum></citerefentry>, | |
1175 | <citerefentry project='die-net'><refentrytitle>debootstrap</refentrytitle><manvolnum>8</manvolnum></citerefentry>, | |
1176 | <citerefentry project='archlinux'><refentrytitle>pacman</refentrytitle><manvolnum>8</manvolnum></citerefentry>, | |
1177 | <citerefentry project='mankier'><refentrytitle>zypper</refentrytitle><manvolnum>8</manvolnum></citerefentry>, | |
1178 | <citerefentry><refentrytitle>systemd.slice</refentrytitle><manvolnum>5</manvolnum></citerefentry>, | |
1179 | <citerefentry><refentrytitle>machinectl</refentrytitle><manvolnum>1</manvolnum></citerefentry>, | |
1180 | <citerefentry project='man-pages'><refentrytitle>btrfs</refentrytitle><manvolnum>8</manvolnum></citerefentry> | |
1181 | </para> | |
1182 | </refsect1> | |
1183 | ||
1184 | </refentry> |