1umount(2) System Calls Manual umount(2)
2
3
4
6 umount, umount2 - unmount filesystem
7
9 Standard C library (libc, -lc)
10
12 #include <sys/mount.h>
13
14 int umount(const char *target);
15 int umount2(const char *target, int flags);
16
18 umount() and umount2() remove the attachment of the (topmost) filesys‐
19 tem mounted on target.
20
21 Appropriate privilege (Linux: the CAP_SYS_ADMIN capability) is required
22 to unmount filesystems.
23
24 Linux 2.1.116 added the umount2() system call, which, like umount(),
25 unmounts a target, but allows additional flags controlling the behavior
26 of the operation:
27
28 MNT_FORCE (since Linux 2.1.116)
29 Ask the filesystem to abort pending requests before attempting
30 the unmount. This may allow the unmount to complete without
31 waiting for an inaccessible server, but could cause data loss.
32 If, after aborting requests, some processes still have active
33 references to the filesystem, the unmount will still fail. As
34 at Linux 4.12, MNT_FORCE is supported only on the following
35 filesystems: 9p (since Linux 2.6.16), ceph (since Linux 2.6.34),
36 cifs (since Linux 2.6.12), fuse (since Linux 2.6.16), lustre
37 (since Linux 3.11), and NFS (since Linux 2.1.116).
38
39 MNT_DETACH (since Linux 2.4.11)
40 Perform a lazy unmount: make the mount unavailable for new ac‐
41 cesses, immediately disconnect the filesystem and all filesys‐
42 tems mounted below it from each other and from the mount table,
43 and actually perform the unmount when the mount ceases to be
44 busy.
45
46 MNT_EXPIRE (since Linux 2.6.8)
47 Mark the mount as expired. If a mount is not currently in use,
48 then an initial call to umount2() with this flag fails with the
49 error EAGAIN, but marks the mount as expired. The mount remains
50 expired as long as it isn't accessed by any process. A second
51 umount2() call specifying MNT_EXPIRE unmounts an expired mount.
52 This flag cannot be specified with either MNT_FORCE or MNT_DE‐
53 TACH.
54
55 UMOUNT_NOFOLLOW (since Linux 2.6.34)
56 Don't dereference target if it is a symbolic link. This flag
57 allows security problems to be avoided in set-user-ID-root pro‐
58 grams that allow unprivileged users to unmount filesystems.
59
61 On success, zero is returned. On error, -1 is returned, and errno is
62 set to indicate the error.
63
65 The error values given below result from filesystem type independent
66 errors. Each filesystem type may have its own special errors and its
67 own special behavior. See the Linux kernel source code for details.
68
69 EAGAIN A call to umount2() specifying MNT_EXPIRE successfully marked an
70 unbusy filesystem as expired.
71
72 EBUSY target could not be unmounted because it is busy.
73
74 EFAULT target points outside the user address space.
75
76 EINVAL target is not a mount point.
77
78 EINVAL target is locked; see mount_namespaces(7).
79
80 EINVAL umount2() was called with MNT_EXPIRE and either MNT_DETACH or
81 MNT_FORCE.
82
83 EINVAL (since Linux 2.6.34)
84 umount2() was called with an invalid flag value in flags.
85
86 ENAMETOOLONG
87 A pathname was longer than MAXPATHLEN.
88
89 ENOENT A pathname was empty or had a nonexistent component.
90
91 ENOMEM The kernel could not allocate a free page to copy filenames or
92 data into.
93
94 EPERM The caller does not have the required privileges.
95
97 Linux.
98
100 MNT_DETACH and MNT_EXPIRE are available since glibc 2.11.
101
102 The original umount() function was called as umount(device) and would
103 return ENOTBLK when called with something other than a block device.
104 In Linux 0.98p4, a call umount(dir) was added, in order to support
105 anonymous devices. In Linux 2.3.99-pre7, the call umount(device) was
106 removed, leaving only umount(dir) (since now devices can be mounted in
107 more than one place, so specifying the device does not suffice).
108
110 umount() and shared mounts
111 Shared mounts cause any mount activity on a mount, including umount()
112 operations, to be forwarded to every shared mount in the peer group and
113 every slave mount of that peer group. This means that umount() of any
114 peer in a set of shared mounts will cause all of its peers to be un‐
115 mounted and all of their slaves to be unmounted as well.
116
117 This propagation of unmount activity can be particularly surprising on
118 systems where every mount is shared by default. On such systems, re‐
119 cursively bind mounting the root directory of the filesystem onto a
120 subdirectory and then later unmounting that subdirectory with MNT_DE‐
121 TACH will cause every mount in the mount namespace to be lazily un‐
122 mounted.
123
124 To ensure umount() does not propagate in this fashion, the mount may be
125 remounted using a mount(2) call with a mount_flags argument that in‐
126 cludes both MS_REC and MS_PRIVATE prior to umount() being called.
127
129 mount(2), mount_namespaces(7), path_resolution(7), mount(8), umount(8)
130
131
132
133Linux man-pages 6.04 2023-03-30 umount(2)