1mount_cachefs(1M) System Administration Commands mount_cachefs(1M)
2
6 mount_cachefs - mount CacheFS file systems
7
9 mount -F cachefs [generic_options] -o backfstype=file_system_type
10 [specific_options]
11 [-O] special mount_point
12
15 The CacheFS-specific version of the mount command mounts a cached file
16 system; if necessary, it NFS-mounts its back file system. It also pro‐
17 vides a number of CacheFS-specific options for controlling the caching
18 process. For more information regarding back file systems, refer to the
19 .
20 mount_cachefs cannot be used with replicated NFS mounts. mount_cachefs
23 creates a pass through when used with an NFS version 4 mount. No
24 caching is performed.
25
27 To mount a CacheFS file system, use the generic mount command with the
28 -F option followed by the argument cachefs.
29 See mount(1M) for a list of supported generic_options.
32 -o specific_options Specify CacheFS file system specific options in
34 a comma-separated list with no intervening spa‐
35 ces.
36 acdirmax=n
38 Specifies that cached attributes are held
40 for no more than n seconds after directory
41 update. After n seconds, all directory
42 information is purged from the cache. The
43 default value is 30 seconds.
44 acdirmin=n
47 Specifies that cached attributes are held
49 for at least n seconds after directory
50 update. After n seconds, CacheFS checks to
51 see if the directory modification time on
52 the back file system has changed. If it has,
53 all information about the directory is
54 purged from the cache and new data is
55 retrieved from the back file system. The
56 default value is 30 seconds.
57 acregmax=n
60 Specifies that cached attributes are held
62 for no more than n seconds after file modi‐
63 fication. After n seconds, all file informa‐
64 tion is purged from the cache. The default
65 value is 30 seconds.
66 acregmin=n
69 Specifies that cached attributes are held
71 for at least n seconds after file modifica‐
72 tion. After n seconds, CacheFS checks to see
73 if the file modification time on the back
74 file system has changed. If it has, all
75 information about the file is purged from
76 the cache and new data is retrieved from the
77 back file system. The default value is 30
78 seconds.
79 actimeo=n
82 Sets acregmin, acregmax, acdirmin, and
84 acdirmax to n.
85 backfstype=file_system_type
88 The file system type of the back file system
90 (can be nfs or hsfs).
91 backpath=path
94 Specifies where the back file system is
96 already mounted. If this argument is not
97 supplied, CacheFS determines a mount point
98 for the back file system. The back file sys‐
99 tem must be read-only.
100 cachedir=directory
103 The name of the cache directory.
105 cacheid=ID
108 ID is a string specifying a particular
110 instance of a cache. If you do not specify a
111 cache ID, CacheFS will construct one.
112 demandconst
115 Verifies cache consistency only when explic‐
117 itly requested, rather than the periodic
118 checking that is done by default. A consis‐
119 tency check is requested by using the -s
120 option of the cfsadmin(1M) command. This
121 option is useful for back file systems that
122 change infrequently, for example, /usr/open‐
123 win. demandconst and noconst are mutually
124 exclusive.
125 local-access
128 Causes the front file system to interpret
130 the mode bits used for access checking
131 instead of having the back file system ver‐
132 ify access permissions. Do not use this
133 argument with secure NFS.
134 noconst
137 Disables cache consistency checking. By
139 default, periodic consistency checking is
140 enabled. Specify noconst only when you know
141 that the back file system will not be modi‐
142 fied. Trying to perform cache consistency
143 check using cfsadmin -s will result in
144 error. demandconst and noconst are mutually
145 exclusive.
146 write-around | non-shared
149 Write modes for CacheFS. The write-around
151 mode (the default) handles writes the same
152 as NFS does; that is, writes are made to the
153 back file system, and the affected file is
154 purged from the cache. You can use the non-
155 shared mode when you are sure that no one
156 else will be writing to the cached file sys‐
157 tem. In this mode, all writes are made to
158 both the front and the back file system, and
159 the file remains in the cache.
160 -O Overlay mount. Allows the filesystem to be
164 mounted over an existing mount point, making the
165 underlying filesystem inaccessible. If a mount
166 is attempted on a pre-existing mount point with‐
167 out setting this flag, mount will fail with the
168 error: mount -F cachefs: mount failed Device
169 busy.
170
173 Example 1 CacheFS-mounting a File System
174 The following example CacheFS-mounts the file system server1:/user2,
177 which is already NFS-mounted on /usr/abc as /xyz.
178 example# mount -F cachefs -o backfstype=nfs,backpath=/usr/abc,
181 cachedir=/cache1 server1:/user2 /xyz
182 The lines similar to the following appear in the /etc/mnttab file after
187 the mount command is executed:
188 server1:/user2 /usr/abc nfs
191 /usr/abc /cache1/xyz cachefs backfstype=nfs
192
196 See attributes(5) for descriptions of the following attributes:
197 ┌─────────────────────────────┬─────────────────────────────┐
202 │ ATTRIBUTE TYPE │ ATTRIBUTE VALUE │
203 ├─────────────────────────────┼─────────────────────────────┤
204 │Availability │SUNWcsu │
205 └─────────────────────────────┴─────────────────────────────┘
206
208 cfsadmin(1M), fsck_cachefs(1M), mount(1M), attributes(5)
209
211 The output for the generic_option -p output is incorrect for cachefs.
212SunOS 5.11 18 Mar 2004 mount_cachefs(1M)