1CHOWN(2)                   Linux Programmer's Manual                  CHOWN(2)
2
3
4

NAME

6       chown, fchown, lchown - change ownership of a file
7

SYNOPSIS

9       #include <unistd.h>
10
11       int chown(const char *path, uid_t owner, gid_t group);
12       int fchown(int fd, uid_t owner, gid_t group);
13       int lchown(const char *path, uid_t owner, gid_t group);
14
15   Feature Test Macro Requirements for glibc (see feature_test_macros(7)):
16
17       fchown(), lchown(): _BSD_SOURCE || _XOPEN_SOURCE >= 500
18

DESCRIPTION

20       These  system  calls  change the owner and group of a file.  The differ
21       only in how the file is specified:
22
23       * chown() changes the ownership of the file specified by path, which is
24         dereferenced if it is a symbolic link.
25
26       * fchown()  changes  the  ownership of the file referred to by the open
27         file descriptor fd.
28
29       * lchown() is like chown(), but does not dereference symbolic links.
30
31       Only a privileged process (Linux: one with  the  CAP_CHOWN  capability)
32       may  change  the  owner  of a file.  The owner of a file may change the
33       group of the file to any group of which that  owner  is  a  member.   A
34       privileged  process  (Linux: with CAP_CHOWN) may change the group arbi‐
35       trarily.
36
37       If the owner or group is specified as -1, then that ID is not changed.
38
39       When the owner or group of an executable file are changed by an unpriv‐
40       ileged  user the S_ISUID and S_ISGID mode bits are cleared.  POSIX does
41       not specify whether this also should happen when root does the chown();
42       the  Linux  behavior  depends on the kernel version.  In case of a non-
43       group-executable file (i.e., one for which the S_IXGRP bit is not  set)
44       the  S_ISGID  bit  indicates mandatory locking, and is not cleared by a
45       chown().
46

RETURN VALUE

48       On success, zero is returned.  On error, -1 is returned, and  errno  is
49       set appropriately.
50

ERRORS

52       Depending  on  the file system, other errors can be returned.  The more
53       general errors for chown() are listed below.
54
55       EACCES Search permission is denied on a component of the  path  prefix.
56              (See also path_resolution(7).)
57
58       EFAULT path points outside your accessible address space.
59
60       ELOOP  Too many symbolic links were encountered in resolving path.
61
62       ENAMETOOLONG
63              path is too long.
64
65       ENOENT The file does not exist.
66
67       ENOMEM Insufficient kernel memory was available.
68
69       ENOTDIR
70              A component of the path prefix is not a directory.
71
72       EPERM  The  calling  process did not have the required permissions (see
73              above) to change owner and/or group.
74
75       EROFS  The named file resides on a read-only file system.
76
77       The general errors for fchown() are listed below:
78
79       EBADF  The descriptor is not valid.
80
81       EIO    A low-level I/O error occurred while modifying the inode.
82
83       ENOENT See above.
84
85       EPERM  See above.
86
87       EROFS  See above.
88

CONFORMING TO

90       4.4BSD, SVr4, POSIX.1-2001.
91
92       The 4.4BSD version can only be used by the superuser (that is, ordinary
93       users cannot give away files).
94

NOTES

96       When  a new file is created (by, for example, open(2) or mkdir(2)), its
97       owner is made the same as the file  system  user  ID  of  the  creating
98       process.   The group of the file depends on a range of factors, includ‐
99       ing the type of file system, the options used to mount the file system,
100       and  whether  or  not the set-group-ID permission bit is enabled on the
101       parent directory.  If the file system supports the -o grpid  (or,  syn‐
102       onymously -o bsdgroups) and -o nogrpid (or, synonymously -o sysvgroups)
103       mount(8) options, then the rules are as follows:
104
105       * If the file system is mounted with -o grpid, then the group of a  new
106         file is made the same as that of the parent directory.
107
108       * If  the  file  system is mounted with -o nogrpid and the set-group-ID
109         bit is disabled on the parent directory, then the group of a new file
110         is made the same as the process's file system GID.
111
112       * If  the  file  system is mounted with -o nogrpid and the set-group-ID
113         bit is enabled on the parent directory, then the group of a new  file
114         is made the same as that of the parent directory.
115
116       As  at Linux 2.6.25, the -o grpid and -o nogrpid mount options are sup‐
117       ported by ext2, ext3, ext4, and XFS.  File systems that  don't  support
118       these mount options follow the -o nogrpid rules.
119
120       The  chown()  semantics  are  deliberately violated on NFS file systems
121       which have UID mapping enabled.  Additionally,  the  semantics  of  all
122       system  calls  which  access  the  file  contents are violated, because
123       chown() may cause immediate access revocation on  already  open  files.
124       Client  side  caching may lead to a delay between the time where owner‐
125       ship have been changed to allow access for a user and  the  time  where
126       the file can actually be accessed by the user on other clients.
127
128       In  versions  of  Linux  prior  to  2.1.81  (and distinct from 2.1.46),
129       chown() did not follow symbolic links.   Since  Linux  2.1.81,  chown()
130       does  follow  symbolic  links,  and there is a new system call lchown()
131       that does not follow symbolic links.  Since Linux 2.1.86, this new call
132       (that  has  the  same  semantics  as  the old chown()) has got the same
133       syscall number, and chown() got the newly introduced number.
134

EXAMPLE

136       The following program changes the ownership of the file  named  in  its
137       second  command-line  argument to the value specified in its first com‐
138       mand-line argument.  The new owner can be specified either as a numeric
139       user  ID,  or  as  a username (which is converted to a user ID by using
140       getpwnam(3) to perform a lookup in the system password file).
141
142       #include <pwd.h>
143       #include <stdio.h>
144       #include <stdlib.h>
145       #include <unistd.h>
146
147       int
148       main(int argc, char *argv[])
149       {
150           uid_t uid;
151           struct passwd *pwd;
152           char *endptr;
153
154           if (argc != 3 || argv[1][0] == '\0') {
155               fprintf(stderr, "%s <owner> <file>\n", argv[0]);
156               exit(EXIT_FAILURE);
157           }
158
159           uid = strtol(argv[1], &endptr, 10);  /* Allow a numeric string */
160
161           if (*endptr != '\0') {         /* Was not pure numeric string */
162               pwd = getpwnam(argv[1]);   /* Try getting UID for username */
163               if (pwd == NULL) {
164                   perror("getpwnam");
165                   exit(EXIT_FAILURE);
166               }
167
168               uid = pwd->pw_uid;
169           }
170
171           if (chown(argv[2], uid, -1) == -1) {
172               perror("chown");
173               exit(EXIT_FAILURE);
174           } /* if */
175
176           exit(EXIT_SUCCESS);
177       } /* main */
178

SEE ALSO

180       chmod(2), fchownat(2), flock(2), path_resolution(7), symlink(7)
181

COLOPHON

183       This page is part of release 3.25 of the Linux  man-pages  project.   A
184       description  of  the project, and information about reporting bugs, can
185       be found at http://www.kernel.org/doc/man-pages/.
186
187
188
189Linux                             2008-06-16                          CHOWN(2)
Impressum