1SETFSUID(2) Linux Programmer's Manual SETFSUID(2)
2
6 setfsuid - set user identity used for filesystem checks
7
9 #include <sys/fsuid.h>
10 int setfsuid(uid_t fsuid);
12
14 The system call setfsuid() changes the value of the caller's filesystem
15 user ID—the user ID that the Linux kernel uses to check for all
16 accesses to the filesystem. Normally, the value of the filesystem user
17 ID will shadow the value of the effective user ID. In fact, whenever
18 the effective user ID is changed, the filesystem user ID will also be
19 changed to the new value of the effective user ID.
20 Explicit calls to setfsuid() and setfsgid(2) are usually used only by
22 programs such as the Linux NFS server that need to change what user and
23 group ID is used for file access without a corresponding change in the
24 real and effective user and group IDs. A change in the normal user IDs
25 for a program such as the NFS server is a security hole that can expose
26 it to unwanted signals. (But see below.)
27 setfsuid() will succeed only if the caller is the superuser or if fsuid
29 matches either the caller's real user ID, effective user ID, saved set-
30 user-ID, or current filesystem user ID.
31
33 On both success and failure, this call returns the previous filesystem
34 user ID of the caller.
35
37 This system call is present in Linux since version 1.2.
38
40 setfsuid() is Linux-specific and should not be used in programs
41 intended to be portable.
42
44 At the time when this system call was introduced, one process could
45 send a signal to another process with the same effective user ID. This
46 meant that if a privileged process changed its effective user ID for
47 the purpose of file permission checking, then it could become vulnera‐
48 ble to receiving signals sent by another (unprivileged) process with
49 the same user ID. The filesystem user ID attribute was thus added to
50 allow a process to change its user ID for the purposes of file permis‐
51 sion checking without at the same time becoming vulnerable to receiving
52 unwanted signals. Since Linux 2.0, signal permission handling is dif‐
53 ferent (see kill(2)), with the result that a process change can change
54 its effective user ID without being vulnerable to receiving signals
55 from unwanted processes. Thus, setfsuid() is nowadays unneeded and
56 should be avoided in new applications (likewise for setfsgid(2)).
57 The original Linux setfsuid() system call supported only 16-bit user
59 IDs. Subsequently, Linux 2.4 added setfsuid32() supporting 32-bit IDs.
60 The glibc setfsuid() wrapper function transparently deals with the
61 variation across kernel versions.
62 C library/kernel differences
64 In glibc 2.15 and earlier, when the wrapper for this system call deter‐
65 mines that the argument can't be passed to the kernel without integer
66 truncation (because the kernel is old and does not support 32-bit user
67 IDs), they will return -1 and set errno to EINVAL without attempting
68 the system call.
69
71 No error indications of any kind are returned to the caller, and the
72 fact that both successful and unsuccessful calls return the same value
73 makes it impossible to directly determine whether the call succeeded or
74 failed. Instead, the caller must resort to looking at the return value
75 from a further call such as setfsuid(-1) (which will always fail), in
76 order to determine if a preceding call to setfsuid() changed the
77 filesystem user ID. At the very least, EPERM should be returned when
78 the call fails (because the caller lacks the CAP_SETUID capability).
79
81 kill(2), setfsgid(2), capabilities(7), credentials(7)
82
84 This page is part of release 4.16 of the Linux man-pages project. A
85 description of the project, information about reporting bugs, and the
86 latest version of this page, can be found at
87 https://www.kernel.org/doc/man-pages/.
88Linux 2017-09-15 SETFSUID(2)