0001 .. SPDX-License-Identifier: GPL-2.0
0002
0003 =================
0004 Automount Support
0005 =================
0006
0007
0008 Support is available for filesystems that wish to do automounting
0009 support (such as kAFS which can be found in fs/afs/ and NFS in
0010 fs/nfs/). This facility includes allowing in-kernel mounts to be
0011 performed and mountpoint degradation to be requested. The latter can
0012 also be requested by userspace.
0013
0014
0015 In-Kernel Automounting
0016 ======================
0017
0018 See section "Mount Traps" of Documentation/filesystems/autofs.rst
0019
0020 Then from userspace, you can just do something like::
0021
0022 [root@andromeda root]# mount -t afs \#root.afs. /afs
0023 [root@andromeda root]# ls /afs
0024 asd cambridge cambridge.redhat.com grand.central.org
0025 [root@andromeda root]# ls /afs/cambridge
0026 afsdoc
0027 [root@andromeda root]# ls /afs/cambridge/afsdoc/
0028 ChangeLog html LICENSE pdf RELNOTES-1.2.2
0029
0030 And then if you look in the mountpoint catalogue, you'll see something like::
0031
0032 [root@andromeda root]# cat /proc/mounts
0033 ...
0034 #root.afs. /afs afs rw 0 0
0035 #root.cell. /afs/cambridge.redhat.com afs rw 0 0
0036 #afsdoc. /afs/cambridge.redhat.com/afsdoc afs rw 0 0
0037
0038
0039 Automatic Mountpoint Expiry
0040 ===========================
0041
0042 Automatic expiration of mountpoints is easy, provided you've mounted the
0043 mountpoint to be expired in the automounting procedure outlined separately.
0044
0045 To do expiration, you need to follow these steps:
0046
0047 (1) Create at least one list off which the vfsmounts to be expired can be
0048 hung.
0049
0050 (2) When a new mountpoint is created in the ->d_automount method, add
0051 the mnt to the list using mnt_set_expiry()::
0052
0053 mnt_set_expiry(newmnt, &afs_vfsmounts);
0054
0055 (3) When you want mountpoints to be expired, call mark_mounts_for_expiry()
0056 with a pointer to this list. This will process the list, marking every
0057 vfsmount thereon for potential expiry on the next call.
0058
0059 If a vfsmount was already flagged for expiry, and if its usage count is 1
0060 (it's only referenced by its parent vfsmount), then it will be deleted
0061 from the namespace and thrown away (effectively unmounted).
0062
0063 It may prove simplest to simply call this at regular intervals, using
0064 some sort of timed event to drive it.
0065
0066 The expiration flag is cleared by calls to mntput. This means that expiration
0067 will only happen on the second expiration request after the last time the
0068 mountpoint was accessed.
0069
0070 If a mountpoint is moved, it gets removed from the expiration list. If a bind
0071 mount is made on an expirable mount, the new vfsmount will not be on the
0072 expiration list and will not expire.
0073
0074 If a namespace is copied, all mountpoints contained therein will be copied,
0075 and the copies of those that are on an expiration list will be added to the
0076 same expiration list.
0077
0078
0079 Userspace Driven Expiry
0080 =======================
0081
0082 As an alternative, it is possible for userspace to request expiry of any
0083 mountpoint (though some will be rejected - the current process's idea of the
0084 rootfs for example). It does this by passing the MNT_EXPIRE flag to
0085 umount(). This flag is considered incompatible with MNT_FORCE and MNT_DETACH.
0086
0087 If the mountpoint in question is in referenced by something other than
0088 umount() or its parent mountpoint, an EBUSY error will be returned and the
0089 mountpoint will not be marked for expiration or unmounted.
0090
0091 If the mountpoint was not already marked for expiry at that time, an EAGAIN
0092 error will be given and it won't be unmounted.
0093
0094 Otherwise if it was already marked and it wasn't referenced, unmounting will
0095 take place as usual.
0096
0097 Again, the expiration flag is cleared every time anything other than umount()
0098 looks at a mountpoint.
