1apptainer(1) apptainer(1)
2
3
4
6 apptainer-instance-start - Start a named instance of the given con‐
7 tainer image
8
9
10
12 apptainer instance start [start options...] [startscript args...]
13
14
15
17 The instance start command allows you to create a new named instance
18 from an
19 existing container image that will begin running in the background.
20 If a
21 startscript is defined in the container metadata the commands in that
22 script
23 will be executed with the instance start command as well. You can op‐
24 tionally
25 pass arguments to startscript
26
27
28 apptainer instance start accepts the following container formats
29
30
31 *.sif Singularity Image Format (SIF). Native to Singular‐
32 ity (3.0+) and Apptainer (v1.0.0+)
33
34
35 *.sqsh SquashFS format. Native to Singularity 2.4+
36
37
38 *.img ext3 format. Native to Singularity versions < 2.4.
39
40
41 directory/ sandbox format. Directory containing a valid root
42 file
43 system and optionally Apptainer meta-data.
44
45
46 instance://* A local running instance of a container. (See the
47 instance
48 command group.)
49
50
51 library://* A SIF container hosted on a Library (no default)
52
53
54 docker://* A Docker/OCI container hosted on Docker Hub or an‐
55 other
56 OCI registry.
57
58
59 shub://* A container hosted on Singularity Hub.
60
61
62 oras://* A SIF container hosted on an OCI registry that sup‐
63 ports
64 the OCI Registry As Storage (ORAS) specification.
65
66
67
69 --add-caps="" a comma separated capability list to add
70
71
72 --allow-setuid[=false] allow setuid binaries in container (root
73 only)
74
75
76 --apply-cgroups="" apply cgroups from file for container processes
77 (root only)
78
79
80 -B, --bind=[] a user-bind path specification. spec has the format
81 src[:dest[:opts]], where src and dest are outside and inside paths. If
82 dest is not given, it is set equal to src. Mount options ('opts') may
83 be specified as 'ro' (read-only) or 'rw' (read/write, which is the de‐
84 fault). Multiple bind paths can be given by a comma separated list.
85
86
87 --boot[=false] execute /sbin/init to boot container (root only)
88
89
90 -e, --cleanenv[=false] clean environment before running container
91
92
93 --compat[=false] apply settings for increased OCI/Docker compati‐
94 bility. Infers --containall, --no-init, --no-umask, --writable-tmpfs.
95
96
97 -c, --contain[=false] use minimal /dev and empty other directories
98 (e.g. /tmp and $HOME) instead of sharing filesystems from your host
99
100
101 -C, --containall[=false] contain not only file systems, but also
102 PID, IPC, and environment
103
104
105 --disable-cache[=false] dont use cache, and dont create cache
106
107
108 --dmtcp-launch="" checkpoint for dmtcp to save container process
109 state to (experimental)
110
111
112 --dmtcp-restart="" checkpoint for dmtcp to use to restart con‐
113 tainer process (experimental)
114
115
116 --dns="" list of DNS server separated by commas to add in re‐
117 solv.conf
118
119
120 --docker-login[=false] login to a Docker Repository interactively
121
122
123 --drop-caps="" a comma separated capability list to drop
124
125
126 --env=[] pass environment variable to contained process
127
128
129 --env-file="" pass environment variables from file to contained
130 process
131
132
133 -f, --fakeroot[=false] run container in new user namespace as uid
134 0
135
136
137 --fusemount=[] A FUSE filesystem mount specification of the form
138 ': ' - where is 'container' or 'host', specifying where the mount will
139 be performed ('container-daemon' or 'host-daemon' will run the FUSE
140 process detached). is the path to the FUSE executable, plus options
141 for the mount. is the location in the container to which the FUSE
142 mount will be attached. E.g. 'container:sshfs 10.0.0.1:/ /sshfs'. Im‐
143 plies --pid.
144
145
146 -h, --help[=false] help for start
147
148
149 -H, --home="/builddir" a home directory specification. spec can
150 either be a src path or src:dest pair. src is the source path of the
151 home directory outside the container and dest overrides the home direc‐
152 tory within the container.
153
154
155 --hostname="" set container hostname
156
157
158 -i, --ipc[=false] run container in a new IPC namespace
159
160
161 --keep-privs[=false] let root user keep privileges in container
162 (root only)
163
164
165 --mount=[] a mount specification e.g. 'type=bind,source=/opt,des‐
166 tination=/hostopt'.
167
168
169 -n, --net[=false] run container in a new network namespace (sets
170 up a bridge network interface by default)
171
172
173 --network="bridge" specify desired network type separated by com‐
174 mas, each network will bring up a dedicated interface inside container
175
176
177 --network-args=[] specify network arguments to pass to CNI plugins
178
179
180 --no-home[=false] do NOT mount users home directory if /home is
181 not the current working directory
182
183
184 --no-https[=false] use http instead of https for docker:// oras://
185 and library:///... URIs
186
187
188 --no-init[=false] do NOT start shim process with --pid
189
190
191 --no-mount=[] disable one or more mount xxx options set in app‐
192 tainer.conf
193
194
195 --no-privs[=false] drop all privileges from root user in con‐
196 tainer)
197
198
199 --no-umask[=false] do not propagate umask to the container, set
200 default 0022 umask
201
202
203 --nv[=false] enable Nvidia support
204
205
206 --nvccli[=false] use nvidia-container-cli for GPU setup (experi‐
207 mental)
208
209
210 -o, --overlay=[] use an overlayFS image for persistent data stor‐
211 age or as read-only layer of container
212
213
214 --passphrase[=false] prompt for an encryption passphrase
215
216
217 --pem-path="" enter an path to a PEM formatted RSA key for an en‐
218 crypted container
219
220
221 --pid-file="" write instance PID to the file with the given name
222
223
224 --rocm[=false] enable experimental Rocm support
225
226
227 -S, --scratch=[] include a scratch directory within the container
228 that is linked to a temporary dir (use -W to force location)
229
230
231 --security=[] enable security features (SELinux, Apparmor, Sec‐
232 comp)
233
234
235 -u, --userns[=false] run container in a new user namespace, allow‐
236 ing Apptainer to run completely unprivileged on recent kernels. This
237 disables some features of Apptainer, for example it only works with
238 sandbox images.
239
240
241 --uts[=false] run container in a new UTS namespace
242
243
244 -W, --workdir="" working directory to be used for /tmp, /var/tmp
245 and $HOME (if -c/--contain was also used)
246
247
248 -w, --writable[=false] by default all Apptainer containers are
249 available as read only. This option makes the file system accessible as
250 read/write.
251
252
253 --writable-tmpfs[=false] makes the file system accessible as read-
254 write with non persistent data (with overlay support only)
255
256
257
259 $ apptainer instance start /tmp/my-sql.sif mysql
260
261 $ apptainer shell instance://mysql
262 Apptainer my-sql.sif> pwd
263 /home/mibauer/mysql
264 Apptainer my-sql.sif> ps
265 PID TTY TIME CMD
266 1 pts/0 00:00:00 sinit
267 2 pts/0 00:00:00 bash
268 3 pts/0 00:00:00 ps
269 Apptainer my-sql.sif>
270
271 $ apptainer instance stop /tmp/my-sql.sif mysql
272 Stopping /tmp/my-sql.sif mysql
273
274
275
276
278 apptainer-instance(1)
279
280
281
283 22-Jun-2022 Auto generated by spf13/cobra
284
285
286
287Auto generated by spf13/cobra Jun 2022 apptainer(1)