man/systemd-nspawn.1

   1 '\" t
   2 .TH "SYSTEMD\-NSPAWN" "1" "" "systemd 208" "systemd-nspawn"
   3 .\" -----------------------------------------------------------------
   4 .\" * Define some portability stuff
   5 .\" -----------------------------------------------------------------
   6 .\" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
   7 .\" http://bugs.debian.org/507673
   8 .\" http://lists.gnu.org/archive/html/groff/2009-02/msg00013.html
   9 .\" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  10 .ie \n(.g .ds Aq \(aq
  11 .el       .ds Aq '
  12 .\" -----------------------------------------------------------------
  13 .\" * set default formatting
  14 .\" -----------------------------------------------------------------
  15 .\" disable hyphenation
  16 .nh
  17 .\" disable justification (adjust text to left margin only)
  18 .ad l
  19 .\" -----------------------------------------------------------------
  20 .\" * MAIN CONTENT STARTS HERE *
  21 .\" -----------------------------------------------------------------
  22 .SH "NAME"
  23 systemd-nspawn \- Spawn a namespace container for debugging, testing and building
  24 .SH "SYNOPSIS"
  25 .HP \w'\fBsystemd\-nspawn\fR\ 'u
  26 \fBsystemd\-nspawn\fR [OPTIONS...] [\fICOMMAND\fR\ [ARGS...]]
  27 .HP \w'\fBsystemd\-nspawn\fR\ 'u
  28 \fBsystemd\-nspawn\fR \-b [OPTIONS...] [ARGS...]
  29 .SH "DESCRIPTION"
  30 .PP
  31 \fBsystemd\-nspawn\fR
  32 may be used to run a command or OS in a light\-weight namespace container\&. In many ways it is similar to
  33 \fBchroot\fR(1), but more powerful since it fully virtualizes the file system hierarchy, as well as the process tree, the various IPC subsystems and the host and domain name\&.
  34 .PP
  35 \fBsystemd\-nspawn\fR
  36 limits access to various kernel interfaces in the container to read\-only, such as
  37 /sys,
  38 /proc/sys
  39 or
  40 /sys/fs/selinux\&. Network interfaces and the system clock may not be changed from within the container\&. Device nodes may not be created\&. The host system cannot be rebooted and kernel modules may not be loaded from within the container\&.
  41 .PP
  42 Note that even though these security precautions are taken
  43 \fBsystemd\-nspawn\fR
  44 is not suitable for secure container setups\&. Many of the security features may be circumvented and are hence primarily useful to avoid accidental changes to the host system from the container\&. The intended use of this program is debugging and testing as well as building of packages, distributions and software involved with boot and systems management\&.
  45 .PP
  46 In contrast to
  47 \fBchroot\fR(1)\ \&\fBsystemd\-nspawn\fR
  48 may be used to boot full Linux\-based operating systems in a container\&.
  49 .PP
  50 Use a tool like
  51 \fByum\fR(8),
  52 \fBdebootstrap\fR(8), or
  53 \fBpacman\fR(8)
  54 to set up an OS directory tree suitable as file system hierarchy for
  55 \fBsystemd\-nspawn\fR
  56 containers\&.
  57 .PP
  58 Note that
  59 \fBsystemd\-nspawn\fR
  60 will mount file systems private to the container to
  61 /dev,
  62 /run
  63 and similar\&. These will not be visible outside of the container, and their contents will be lost when the container exits\&.
  64 .PP
  65 Note that running two
  66 \fBsystemd\-nspawn\fR
  67 containers from the same directory tree will not make processes in them see each other\&. The PID namespace separation of the two containers is complete and the containers will share very few runtime objects except for the underlying file system\&. It is however possible to enter an existing container, see
  68 Example 4
  69 below\&.
  70 .PP
  71 \fBsystemd\-nspawn\fR
  72 implements the
  73 \m[blue]\fBContainer Interface\fR\m[]\&\s-2\u[1]\d\s+2
  74 specification\&.
  75 .PP
  76 As a safety check
  77 \fBsystemd\-nspawn\fR
  78 will verify the existence of
  79 /etc/os\-release
  80 in the container tree before starting the container (see
  81 \fBos-release\fR(5))\&. It might be necessary to add this file to the container tree manually if the OS of the container is too old to contain this file out\-of\-the\-box\&.
  82 .SH "INCOMPATIBILITY WITH AUDITING"
  83 .PP
  84 Note that the kernel auditing subsystem is currently broken when used together with containers\&. We hence recommend turning it off entirely by booting with
  85 "audit=0"
  86 on the kernel command line, or by turning it off at kernel build time\&. If auditing is enabled in the kernel, operating systems booted in an nspawn container might refuse log\-in attempts\&.
  87 .SH "OPTIONS"
  88 .PP
  89 If option
  90 \fB\-b\fR
  91 is specified, the arguments are used as arguments for the init binary\&. Otherwise,
  92 \fICOMMAND\fR
  93 specifies the program to launch in the container, and the remaining arguments are used as arguments for this program\&. If
  94 \fB\-b\fR
  95 is not used and no arguments are specifed, a shell is launched in the container\&.
  96 .PP
  97 The following options are understood:
  98 .PP
  99 \fB\-h\fR, \fB\-\-help\fR
 100 .RS 4
 101 Prints a short help text and exits\&.
 102 .RE
 103 .PP
 104 \fB\-\-version\fR
 105 .RS 4
 106 Prints a version string and exits\&.
 107 .RE
 108 .PP
 109 \fB\-D\fR, \fB\-\-directory=\fR
 110 .RS 4
 111 Directory to use as file system root for the namespace container\&. If omitted, the current directory will be used\&.
 112 .RE
 113 .PP
 114 \fB\-b\fR, \fB\-\-boot\fR
 115 .RS 4
 116 Automatically search for an init binary and invoke it instead of a shell or a user supplied program\&. If this option is used, arguments specified on the command line are used as arguments for the init binary\&.
 117 .RE
 118 .PP
 119 \fB\-u\fR, \fB\-\-user=\fR
 120 .RS 4
 121 Run the command under specified user, create home directory and cd into it\&. As rest of systemd\-nspawn, this is not the security feature and limits against accidental changes only\&.
 122 .RE
 123 .PP
 124 \fB\-M\fR, \fB\-\-machine=\fR
 125 .RS 4
 126 Sets the machine name for this container\&. This name may be used to identify this container on the host, and is used to initialize the container\*(Aqs hostname (which the container can choose to override, however)\&. If not specified, the last component of the root directory of the container is used\&.
 127 .RE
 128 .PP
 129 \fB\-\-slice=\fR
 130 .RS 4
 131 Make the container part of the specified slice, instead of the
 132 machine\&.slice\&.
 133 .RE
 134 .PP
 135 \fB\-\-uuid=\fR
 136 .RS 4
 137 Set the specified UUID for the container\&. The init system will initialize
 138 /etc/machine\-id
 139 from this if this file is not set yet\&.
 140 .RE
 141 .PP
 142 \fB\-\-private\-network\fR
 143 .RS 4
 144 Turn off networking in the container\&. This makes all network interfaces unavailable in the container, with the exception of the loopback device\&.
 145 .RE
 146 .PP
 147 \fB\-\-read\-only\fR
 148 .RS 4
 149 Mount the root file system read\-only for the container\&.
 150 .RE
 151 .PP
 152 \fB\-\-capability=\fR
 153 .RS 4
 154 List one or more additional capabilities to grant the container\&. Takes a comma\-separated list of capability names, see
 155 \fBcapabilities\fR(7)
 156 for more information\&. Note that the following capabilities will be granted in any way: CAP_CHOWN, CAP_DAC_OVERRIDE, CAP_DAC_READ_SEARCH, CAP_FOWNER, CAP_FSETID, CAP_IPC_OWNER, CAP_KILL, CAP_LEASE, CAP_LINUX_IMMUTABLE, CAP_NET_BIND_SERVICE, CAP_NET_BROADCAST, CAP_NET_RAW, CAP_SETGID, CAP_SETFCAP, CAP_SETPCAP, CAP_SETUID, CAP_SYS_ADMIN, CAP_SYS_CHROOT, CAP_SYS_NICE, CAP_SYS_PTRACE, CAP_SYS_TTY_CONFIG, CAP_SYS_RESOURCE, CAP_SYS_BOOT, CAP_AUDIT_WRITE, CAP_AUDIT_CONTROL\&.
 157 .RE
 158 .PP
 159 \fB\-\-link\-journal=\fR
 160 .RS 4
 161 Control whether the container\*(Aqs journal shall be made visible to the host system\&. If enabled, allows viewing the container\*(Aqs journal files from the host (but not vice versa)\&. Takes one of
 162 "no",
 163 "host",
 164 "guest",
 165 "auto"\&. If
 166 "no", the journal is not linked\&. If
 167 "host", the journal files are stored on the host file system (beneath
 168 /var/log/journal/\fImachine\-id\fR) and the subdirectory is bind\-mounted into the container at the same location\&. If
 169 "guest", the journal files are stored on the guest file system (beneath
 170 /var/log/journal/\fImachine\-id\fR) and the subdirectory is symlinked into the host at the same location\&. If
 171 "auto"
 172 (the default), and the right subdirectory of
 173 /var/log/journal
 174 exists, it will be bind mounted into the container\&. If the subdirectory does not exist, no linking is performed\&. Effectively, booting a container once with
 175 "guest"
 176 or
 177 "host"
 178 will link the journal persistently if further on the default of
 179 "auto"
 180 is used\&.
 181 .RE
 182 .PP
 183 \fB\-j\fR
 184 .RS 4
 185 Equivalent to
 186 \fB\-\-link\-journal=guest\fR\&.
 187 .RE
 188 .PP
 189 \fB\-\-bind=\fR, \fB\-\-bind\-ro=\fR
 190 .RS 4
 191 Bind mount a file or directory from the host into the container\&. Either takes a path argument \-\- in which case the specified path will be mounted from the host to the same path in the container \-\-, or a colon\-separated pair of paths \-\- in which case the first specified path is the source in the host, and the second path is the destination in the container\&. The
 192 \fB\-\-bind\-ro=\fR
 193 option creates read\-only bind mount\&.
 194 .RE
 195 .SH "EXAMPLE 1"
 196 .sp
 197 .if n \{\
 198 .RS 4
 199 .\}
 200 .nf
 201 # yum \-y \-\-releasever=19 \-\-nogpg \-\-installroot=/srv/mycontainer \-\-disablerepo=\*(Aq*\*(Aq \-\-enablerepo=fedora install systemd passwd yum fedora\-release vim\-minimal
 202 # systemd\-nspawn \-bD /srv/mycontainer
 203 .fi
 204 .if n \{\
 205 .RE
 206 .\}
 207 .PP
 208 This installs a minimal Fedora distribution into the directory
 209 /srv/mycontainer/
 210 and then boots an OS in a namespace container in it\&.
 211 .SH "EXAMPLE 2"
 212 .sp
 213 .if n \{\
 214 .RS 4
 215 .\}
 216 .nf
 217 # debootstrap \-\-arch=amd64 unstable ~/debian\-tree/
 218 # systemd\-nspawn \-D ~/debian\-tree/
 219 .fi
 220 .if n \{\
 221 .RE
 222 .\}
 223 .PP
 224 This installs a minimal Debian unstable distribution into the directory
 225 ~/debian\-tree/
 226 and then spawns a shell in a namespace container in it\&.
 227 .SH "EXAMPLE 3"
 228 .sp
 229 .if n \{\
 230 .RS 4
 231 .\}
 232 .nf
 233 # pacstrap \-c \-d ~/arch\-tree/ base
 234 # systemd\-nspawn \-bD ~/arch\-tree/
 235 .fi
 236 .if n \{\
 237 .RE
 238 .\}
 239 .PP
 240 This installs a mimimal Arch Linux distribution into the directory
 241 ~/arch\-tree/
 242 and then boots an OS in a namespace container in it\&.
 243 .SH "EXAMPLE 4"
 244 .PP
 245 To enter the container, PID of one of the processes sharing the new namespaces must be used\&.
 246 \fBsystemd\-nspawn\fR
 247 prints the PID (as viewed from the outside) of the launched process, and it can be used to enter the container\&.
 248 .sp
 249 .if n \{\
 250 .RS 4
 251 .\}
 252 .nf
 253 # nsenter \-m \-u \-i \-n \-p \-t $PID
 254 .fi
 255 .if n \{\
 256 .RE
 257 .\}
 258 .PP
 259 \fBnsenter\fR(1)
 260 is part of
 261 \m[blue]\fButil\-linux\fR\m[]\&\s-2\u[2]\d\s+2\&. Kernel support for entering namespaces was added in Linux 3\&.8\&.
 262 .SH "EXIT STATUS"
 263 .PP
 264 The exit code of the program executed in the container is returned\&.
 265 .SH "SEE ALSO"
 266 .PP
 267 \fBsystemd\fR(1),
 268 \fBchroot\fR(1),
 269 \fBunshare\fR(1),
 270 \fByum\fR(8),
 271 \fBdebootstrap\fR(8),
 272 \fBpacman\fR(8),
 273 \fBsystemd.slice\fR(5)
 274 .SH "NOTES"
 275 .IP " 1." 4
 276 Container Interface
 277 .RS 4
 278 \%http://www.freedesktop.org/wiki/Software/systemd/ContainerInterface
 279 .RE
 280 .IP " 2." 4
 281 util-linux
 282 .RS 4
 283 \%https://github.com/karelzak/util-linux
 284 .RE