| blob | feff5a0b0d96b87f7047bfeac7b7bbe798737645 |
1 .\" Copyright (C) 2006, Janak Desai <janak@us.ibm.com>
2 .\" and Copyright (C) 2006, 2012 Michael Kerrisk <mtk.manpages@gmail.com>
3 .\"
4 .\" %%%LICENSE_START(GPL_NOVERSION_ONELINE)
5 .\" Licensed under the GPL
6 .\" %%%LICENSE_END
7 .\"
8 .\" Patch Justification:
9 .\" unshare system call is needed to implement, using PAM,
10 .\" per-security_context and/or per-user namespace to provide
11 .\" polyinstantiated directories. Using unshare and bind mounts, a
12 .\" PAM module can create private namespace with appropriate
13 .\" directories(based on user's security context) bind mounted on
14 .\" public directories such as /tmp, thus providing an instance of
15 .\" /tmp that is based on user's security context. Without the
16 .\" unshare system call, namespace separation can only be achieved
17 .\" by clone, which would require porting and maintaining all commands
18 .\" such as login, and su, that establish a user session.
19 .\"
20 .TH UNSHARE 2 2017-05-03 "Linux" "Linux Programmer's Manual"
21 .SH NAME
22 unshare \- disassociate parts of the process execution context
23 .SH SYNOPSIS
24 .nf
25 .B #define _GNU_SOURCE
26 .B #include <sched.h>
27 .sp
28 .BI "int unshare(int " flags );
29 .fi
30 .SH DESCRIPTION
31 .BR unshare ()
32 allows a process (or thread) to disassociate parts of its execution
33 context that are currently being shared with other processes (or threads).
34 Part of the execution context, such as the mount namespace, is shared
35 implicitly when a new process is created using
36 .BR fork (2)
37 or
38 .BR vfork (2),
39 while other parts, such as virtual memory, may be
40 shared by explicit request when creating a process or thread using
41 .BR clone (2).
43 The main use of
44 .BR unshare ()
45 is to allow a process to control its
46 shared execution context without creating a new process.
48 The
49 .I flags
50 argument is a bit mask that specifies which parts of
51 the execution context should be unshared.
52 This argument is specified by ORing together zero or more
53 of the following constants:
54 .TP
55 .B CLONE_FILES
56 Reverse the effect of the
57 .BR clone (2)
58 .B CLONE_FILES
59 flag.
60 Unshare the file descriptor table, so that the calling process
61 no longer shares its file descriptors with any other process.
62 .TP
63 .B CLONE_FS
64 Reverse the effect of the
65 .BR clone (2)
66 .B CLONE_FS
67 flag.
68 Unshare filesystem attributes, so that the calling process
69 no longer shares its root directory
70 .RB ( chroot (2)),
71 current directory
72 .RB ( chdir (2)),
73 or umask
74 .RB ( umask (2))
75 attributes with any other process.
76 .TP
77 .BR CLONE_NEWCGROUP " (since Linux 4.6)"
78 This flag has the same effect as the
79 .BR clone (2)
80 .B CLONE_NEWCGROUP
81 flag.
82 Unshare the cgroup namespace.
83 Use of
84 .BR CLONE_NEWCGROUP
85 requires the
86 .BR CAP_SYS_ADMIN
87 capability.
88 .TP
89 .BR CLONE_NEWIPC " (since Linux 2.6.19)"
90 This flag has the same effect as the
91 .BR clone (2)
92 .B CLONE_NEWIPC
93 flag.
94 Unshare the IPC namespace,
95 so that the calling process has a private copy of the
96 IPC namespace which is not shared with any other process.
97 Specifying this flag automatically implies
98 .BR CLONE_SYSVSEM
99 as well.
100 Use of
101 .BR CLONE_NEWIPC
102 requires the
103 .BR CAP_SYS_ADMIN
104 capability.
105 .TP
106 .BR CLONE_NEWNET " (since Linux 2.6.24)"
107 This flag has the same effect as the
108 .BR clone (2)
109 .B CLONE_NEWNET
110 flag.
111 Unshare the network namespace,
112 so that the calling process is moved into a
113 new network namespace which is not shared
114 with any previously existing process.
115 Use of
116 .BR CLONE_NEWNET
117 requires the
118 .BR CAP_SYS_ADMIN
119 capability.
120 .TP
121 .B CLONE_NEWNS
122 .\" These flag name are inconsistent:
123 .\" CLONE_NEWNS does the same thing in clone(), but CLONE_VM,
124 .\" CLONE_FS, and CLONE_FILES reverse the action of the clone()
125 .\" flags of the same name.
126 This flag has the same effect as the
127 .BR clone (2)
128 .B CLONE_NEWNS
129 flag.
130 Unshare the mount namespace,
131 so that the calling process has a private copy of
132 its namespace which is not shared with any other process.
133 Specifying this flag automatically implies
134 .B CLONE_FS
135 as well.
136 Use of
137 .BR CLONE_NEWNS
138 requires the
139 .BR CAP_SYS_ADMIN
140 capability.
141 For further information, see
142 .BR mount_namespaces (7).
143 .TP
144 .BR CLONE_NEWPID " (since Linux 3.8)"
145 This flag has the same effect as the
146 .BR clone (2)
147 .B CLONE_NEWPID
148 flag.
149 Unshare the PID namespace,
150 so that the calling process has a new PID namespace for its children
151 which is not shared with any previously existing process.
152 The calling process is
153 .I not
154 moved into the new namespace.
155 The first child created by the calling process will have
156 the process ID 1 and will assume the role of
157 .BR init (1)
158 in the new namespace.
159 .BR CLONE_NEWPID
160 automatically implies
161 .BR CLONE_THREAD
162 as well.
163 Use of
164 .BR CLONE_NEWPID
165 requires the
166 .BR CAP_SYS_ADMIN
167 capability.
168 For further information, see
169 .BR pid_namespaces (7).
170 .TP
171 .BR CLONE_NEWUSER " (since Linux 3.8)"
172 This flag has the same effect as the
173 .BR clone (2)
174 .B CLONE_NEWUSER
175 flag.
176 Unshare the user namespace,
177 so that the calling process is moved into a new user namespace
178 which is not shared with any previously existing process.
179 As with the child process created by
180 .BR clone (2)
181 with the
182 .B CLONE_NEWUSER
183 flag, the caller obtains a full set of capabilities in the new namespace.
184 .IP
185 .BR CLONE_NEWUSER
186 requires that the calling process is not threaded; specifying
187 .BR CLONE_NEWUSER
188 automatically implies
189 .BR CLONE_THREAD .
190 Since Linux 3.9,
191 .\" commit e66eded8309ebf679d3d3c1f5820d1f2ca332c71
192 .\" https://lwn.net/Articles/543273/
193 .BR CLONE_NEWUSER
194 also automatically implies
195 .BR CLONE_FS .
196 .BR CLONE_NEWUSER
197 requires that the user ID and group ID
198 of the calling process are mapped to user IDs and group IDs in the
199 user namespace of the calling process at the time of the call.
201 For further information on user namespaces, see
202 .BR user_namespaces (7).
203 .TP
204 .BR CLONE_NEWUTS " (since Linux 2.6.19)"
205 This flag has the same effect as the
206 .BR clone (2)
207 .B CLONE_NEWUTS
208 flag.
209 Unshare the UTS IPC namespace,
210 so that the calling process has a private copy of the
211 UTS namespace which is not shared with any other process.
212 Use of
213 .BR CLONE_NEWUTS
214 requires the
215 .BR CAP_SYS_ADMIN
216 capability.
217 .TP
218 .BR CLONE_SYSVSEM " (since Linux 2.6.26)
219 .\" commit 9edff4ab1f8d82675277a04e359d0ed8bf14a7b7
220 This flag reverses the effect of the
221 .BR clone (2)
222 .B CLONE_SYSVSEM
223 flag.
224 Unshare System\ V semaphore adjustment
225 .RI ( semadj )
226 values,
227 so that the calling process has a new empty
228 .I semadj
229 list that is not shared with any other process.
230 If this is the last process that has a reference to the process's current
231 .I semadj
232 list, then the adjustments in that list are applied
233 to the corresponding semaphores, as described in
234 .BR semop (2).
235 .\" CLONE_NEWNS If CLONE_SIGHAND is set and signals are also being shared
236 .\" (i.e., current->signal->count > 1), force CLONE_THREAD.
237 .PP
238 In addition,
239 .BR CLONE_THREAD ,
240 .BR CLONE_SIGHAND ,
241 and
242 .BR CLONE_VM
243 can be specified in
244 .I flags
245 if the caller is single threaded (i.e., it is not sharing
246 its address space with another process or thread).
247 In this case, these flags have no effect.
248 (Note also that specifying
249 .BR CLONE_THREAD
250 automatically implies
251 .BR CLONE_VM ,
252 and specifying
253 .BR CLONE_VM
254 automatically implies
255 .BR CLONE_SIGHAND .)
256 .\" As at 3.9, the following forced implications also apply,
257 .\" although the relevant flags are not yet implemented.
258 .\" If CLONE_THREAD is set force CLONE_VM.
259 .\" If CLONE_VM is set, force CLONE_SIGHAND.
260 .\"
261 If the process is multithreaded, then
262 the use of these flags results in an error.
263 .\" See kernel/fork.c::check_unshare_flags()
264 .PP
265 If
266 .I flags
267 is specified as zero, then
268 .BR unshare ()
269 is a no-op;
270 no changes are made to the calling process's execution context.
271 .SH RETURN VALUE
272 On success, zero returned.
273 On failure, \-1 is returned and
274 .I errno
275 is set to indicate the error.
276 .SH ERRORS
277 .TP
278 .B EINVAL
279 An invalid bit was specified in
280 .IR flags .
281 .TP
282 .B EINVAL
283 .BR CLONE_THREAD ,
284 .BR CLONE_SIGHAND ,
285 or
286 .BR CLONE_VM
287 was specified in
288 .IR flags ,
289 and the caller is multithreaded.
290 .TP
291 .B ENOMEM
292 Cannot allocate sufficient memory to copy parts of caller's
293 context that need to be unshared.
294 .TP
295 .BR ENOSPC " (since Linux 3.7)"
296 .\" commit f2302505775fd13ba93f034206f1e2a587017929
297 .B CLONE_NEWPID
298 was specified in flags,
299 but the limit on the nesting depth of PID namespaces
300 would have been exceeded; see
301 .BR pid_namespaces (7).
302 .TP
303 .BR ENOSPC " (since Linux 4.9; beforehand " EUSERS )
304 .B CLONE_NEWUSER
305 was specified in
306 .IR flags ,
307 and the call would cause the limit on the number of
308 nested user namespaces to be exceeded.
309 See
310 .BR user_namespaces (7).
312 From Linux 3.11 to Linux 4.8, the error diagnosed in this case was
313 .BR EUSERS .
314 .TP
315 .BR ENOSPC " (since Linux 4.9)"
316 One of the values in
317 .I flags
318 specified the creation of a new user namespace,
319 but doing so would have caused the limit defined by the corresponding file in
320 .IR /proc/sys/user
321 to be exceeded.
322 For further details, see
323 .BR namespaces (7).
324 .TP
325 .B EPERM
326 The calling process did not have the required privileges for this operation.
327 .TP
328 .B EPERM
329 .BR CLONE_NEWUSER
330 was specified in
331 .IR flags ,
332 but either the effective user ID or the effective group ID of the caller
333 does not have a mapping in the parent namespace (see
334 .BR user_namespaces (7)).
335 .TP
336 .BR EPERM " (since Linux 3.9)"
337 .\" commit 3151527ee007b73a0ebd296010f1c0454a919c7d
338 .B CLONE_NEWUSER
339 was specified in
340 .I flags
341 and the caller is in a chroot environment
342 .\" FIXME What is the rationale for this restriction?
343 (i.e., the caller's root directory does not match the root directory
344 of the mount namespace in which it resides).
345 .TP
346 .BR EUSERS " (since Linux 3.11)"
347 .B CLONE_NEWUSER
348 was specified in
349 .IR flags ,
350 and the limit on the number of nested user namespaces would be exceeded.
351 See the discussion of the
352 .BR ENOSPC
353 error above.
354 .SH VERSIONS
355 The
356 .BR unshare ()
357 system call was added to Linux in kernel 2.6.16.
358 .SH CONFORMING TO
359 The
360 .BR unshare ()
361 system call is Linux-specific.
362 .SH NOTES
363 Not all of the process attributes that can be shared when
364 a new process is created using
365 .BR clone (2)
366 can be unshared using
367 .BR unshare ().
368 In particular, as at kernel 3.8,
369 .\" FIXME all of the following needs to be reviewed for the current kernel
370 .BR unshare ()
371 does not implement flags that reverse the effects of
372 .BR CLONE_SIGHAND ,
373 .\" However, we can do unshare(CLONE_SIGHAND) if CLONE_SIGHAND
374 .\" was not specified when doing clone(); i.e., unsharing
375 .\" signal handlers is permitted if we are not actually
376 .\" sharing signal handlers. mtk
377 .BR CLONE_THREAD ,
378 or
379 .BR CLONE_VM .
380 .\" However, we can do unshare(CLONE_VM) if CLONE_VM
381 .\" was not specified when doing clone(); i.e., unsharing
382 .\" virtual memory is permitted if we are not actually
383 .\" sharing virtual memory. mtk
384 Such functionality may be added in the future, if required.
385 .\"
386 .\"9) Future Work
387 .\"--------------
388 .\"The current implementation of unshare does not allow unsharing of
389 .\"signals and signal handlers. Signals are complex to begin with and
390 .\"to unshare signals and/or signal handlers of a currently running
391 .\"process is even more complex. If in the future there is a specific
392 .\"need to allow unsharing of signals and/or signal handlers, it can
393 .\"be incrementally added to unshare without affecting legacy
394 .\"applications using unshare.
395 .\"
396 .SH EXAMPLE
397 The program below provides a simple implementation of the
398 .BR unshare (1)
399 command, which unshares one or more namespaces and executes the
400 command supplied in its command-line arguments.
401 Here's an example of the use of this program,
402 running a shell in a new mount namespace,
403 and verifying that the original shell and the
404 new shell are in separate mount namespaces:
405 .in +4n
406 .nf
408 $ \fBreadlink /proc/$$/ns/mnt\fP
409 mnt:[4026531840]
410 $ \fBsudo ./unshare -m /bin/bash\fP
411 [sudo] password for cecilia:
412 # \fBreadlink /proc/$$/ns/mnt\fP
413 mnt:[4026532325]
414 .fi
415 .in
417 The differing output of the two
418 .BR readlink (1)
419 commands shows that the two shells are in different mount namespaces.
420 .SS Program source
421 \&
422 .nf
423 /* unshare.c
425 A simple implementation of the unshare(1) command: unshare
426 namespaces and execute a command.
427 */
428 #define _GNU_SOURCE
429 #include <sched.h>
430 #include <unistd.h>
431 #include <stdlib.h>
432 #include <stdio.h>
434 /* A simple error\-handling function: print an error message based
435 on the value in \(aqerrno\(aq and terminate the calling process */
437 #define errExit(msg) do { perror(msg); exit(EXIT_FAILURE); \\
438 } while (0)
440 static void
441 usage(char *pname)
442 {
443 fprintf(stderr, "Usage: %s [options] program [arg...]\\n", pname);
444 fprintf(stderr, "Options can be:\\n");
445 fprintf(stderr, " \-i unshare IPC namespace\\n");
446 fprintf(stderr, " \-m unshare mount namespace\\n");
447 fprintf(stderr, " \-n unshare network namespace\\n");
448 fprintf(stderr, " \-p unshare PID namespace\\n");
449 fprintf(stderr, " \-u unshare UTS namespace\\n");
450 fprintf(stderr, " \-U unshare user namespace\\n");
451 exit(EXIT_FAILURE);
452 }
454 int
455 main(int argc, char *argv[])
456 {
457 int flags, opt;
459 flags = 0;
461 while ((opt = getopt(argc, argv, "imnpuU")) != \-1) {
462 switch (opt) {
463 case \(aqi\(aq: flags |= CLONE_NEWIPC; break;
464 case \(aqm\(aq: flags |= CLONE_NEWNS; break;
465 case \(aqn\(aq: flags |= CLONE_NEWNET; break;
466 case \(aqp\(aq: flags |= CLONE_NEWPID; break;
467 case \(aqu\(aq: flags |= CLONE_NEWUTS; break;
468 case \(aqU\(aq: flags |= CLONE_NEWUSER; break;
469 default: usage(argv[0]);
470 }
471 }
473 if (optind >= argc)
474 usage(argv[0]);
476 if (unshare(flags) == \-1)
477 errExit("unshare");
479 execvp(argv[optind], &argv[optind]);
480 errExit("execvp");
481 }
482 .fi
483 .SH SEE ALSO
484 .BR unshare (1),
485 .BR clone (2),
486 .BR fork (2),
487 .BR kcmp (2),
488 .BR setns (2),
489 .BR vfork (2),
490 .BR namespaces (7)
492 .I Documentation/unshare.txt
493 in the Linux kernel source tree