1CHOWN(2) Linux Programmer's Manual CHOWN(2)
2
6 chown, fchown, lchown - change ownership of a file
7
9 #include <unistd.h>
10 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 Feature Test Macro Requirements for glibc (see feature_test_macros(7)):
16 fchown(), lchown():
18 _BSD_SOURCE || _XOPEN_SOURCE >= 500 ||
19 _XOPEN_SOURCE && _XOPEN_SOURCE_EXTENDED
20 || /* Since glibc 2.12: */ _POSIX_C_SOURCE >= 200809L
21
23 These system calls change the owner and group of a file. They differ
24 only in how the file is specified:
25 * chown() changes the ownership of the file specified by path, which is
27 dereferenced if it is a symbolic link.
28 * fchown() changes the ownership of the file referred to by the open
30 file descriptor fd.
31 * lchown() is like chown(), but does not dereference symbolic links.
33 Only a privileged process (Linux: one with the CAP_CHOWN capability)
35 may change the owner of a file. The owner of a file may change the
36 group of the file to any group of which that owner is a member. A
37 privileged process (Linux: with CAP_CHOWN) may change the group arbi‐
38 trarily.
39 If the owner or group is specified as -1, then that ID is not changed.
41 When the owner or group of an executable file are changed by an unpriv‐
43 ileged user the S_ISUID and S_ISGID mode bits are cleared. POSIX does
44 not specify whether this also should happen when root does the chown();
45 the Linux behavior depends on the kernel version. In case of a non-
46 group-executable file (i.e., one for which the S_IXGRP bit is not set)
47 the S_ISGID bit indicates mandatory locking, and is not cleared by a
48 chown().
49
51 On success, zero is returned. On error, -1 is returned, and errno is
52 set appropriately.
53
55 Depending on the file system, other errors can be returned. The more
56 general errors for chown() are listed below.
57 EACCES Search permission is denied on a component of the path prefix.
59 (See also path_resolution(7).)
60 EFAULT path points outside your accessible address space.
62 ELOOP Too many symbolic links were encountered in resolving path.
64 ENAMETOOLONG
66 path is too long.
67 ENOENT The file does not exist.
69 ENOMEM Insufficient kernel memory was available.
71 ENOTDIR
73 A component of the path prefix is not a directory.
74 EPERM The calling process did not have the required permissions (see
76 above) to change owner and/or group.
77 EROFS The named file resides on a read-only file system.
79 The general errors for fchown() are listed below:
81 EBADF The descriptor is not valid.
83 EIO A low-level I/O error occurred while modifying the inode.
85 ENOENT See above.
87 EPERM See above.
89 EROFS See above.
91
93 4.4BSD, SVr4, POSIX.1-2001.
94 The 4.4BSD version can be used only by the superuser (that is, ordinary
96 users cannot give away files).
97
99 The original Linux chown(), fchown(), and lchown() system calls sup‐
100 ported only 16-bit user and group IDs. Subsequently, Linux 2.4 added
101 chown32(), fchown32(), and lchown32(), supporting 32-bit IDs. The
102 glibc chown(), fchown(), and lchown() wrapper functions transparently
103 deal with the variations across kernel versions.
104 When a new file is created (by, for example, open(2) or mkdir(2)), its
106 owner is made the same as the file system user ID of the creating
107 process. The group of the file depends on a range of factors, includ‐
108 ing the type of file system, the options used to mount the file system,
109 and whether or not the set-group-ID permission bit is enabled on the
110 parent directory. If the file system supports the -o grpid (or, syn‐
111 onymously -o bsdgroups) and -o nogrpid (or, synonymously -o sysvgroups)
112 mount(8) options, then the rules are as follows:
113 * If the file system is mounted with -o grpid, then the group of a new
115 file is made the same as that of the parent directory.
116 * If the file system is mounted with -o nogrpid and the set-group-ID
118 bit is disabled on the parent directory, then the group of a new file
119 is made the same as the process's file system GID.
120 * If the file system is mounted with -o nogrpid and the set-group-ID
122 bit is enabled on the parent directory, then the group of a new file
123 is made the same as that of the parent directory.
124 As at Linux 2.6.25, the -o grpid and -o nogrpid mount options are sup‐
126 ported by ext2, ext3, ext4, and XFS. File systems that don't support
127 these mount options follow the -o nogrpid rules.
128 The chown() semantics are deliberately violated on NFS file systems
130 which have UID mapping enabled. Additionally, the semantics of all
131 system calls which access the file contents are violated, because
132 chown() may cause immediate access revocation on already open files.
133 Client side caching may lead to a delay between the time where owner‐
134 ship have been changed to allow access for a user and the time where
135 the file can actually be accessed by the user on other clients.
136 In versions of Linux prior to 2.1.81 (and distinct from 2.1.46),
138 chown() did not follow symbolic links. Since Linux 2.1.81, chown()
139 does follow symbolic links, and there is a new system call lchown()
140 that does not follow symbolic links. Since Linux 2.1.86, this new call
141 (that has the same semantics as the old chown()) has got the same
142 syscall number, and chown() got the newly introduced number.
143
145 The following program changes the ownership of the file named in its
146 second command-line argument to the value specified in its first com‐
147 mand-line argument. The new owner can be specified either as a numeric
148 user ID, or as a username (which is converted to a user ID by using
149 getpwnam(3) to perform a lookup in the system password file).
150 #include <pwd.h>
152 #include <stdio.h>
153 #include <stdlib.h>
154 #include <unistd.h>
155 int
157 main(int argc, char *argv[])
158 {
159 uid_t uid;
160 struct passwd *pwd;
161 char *endptr;
162 if (argc != 3 || argv[1][0] == '\0') {
164 fprintf(stderr, "%s <owner> <file>\n", argv[0]);
165 exit(EXIT_FAILURE);
166 }
167 uid = strtol(argv[1], &endptr, 10); /* Allow a numeric string */
169 if (*endptr != '\0') { /* Was not pure numeric string */
171 pwd = getpwnam(argv[1]); /* Try getting UID for username */
172 if (pwd == NULL) {
173 perror("getpwnam");
174 exit(EXIT_FAILURE);
175 }
176 uid = pwd->pw_uid;
178 }
179 if (chown(argv[2], uid, -1) == -1) {
181 perror("chown");
182 exit(EXIT_FAILURE);
183 }
184 exit(EXIT_SUCCESS);
186 }
187
189 chmod(2), fchownat(2), flock(2), path_resolution(7), symlink(7)
190
192 This page is part of release 3.53 of the Linux man-pages project. A
193 description of the project, and information about reporting bugs, can
194 be found at http://www.kernel.org/doc/man-pages/.
195Linux 2010-11-22 CHOWN(2)