1AUTO.MASTER(5) File Formats Manual AUTO.MASTER(5)
2
3
4
6 auto.master - Master Map for automounter consulted by autofs
7
9 The auto.master map is consulted to set up automount managed mount
10 points when the autofs(8) script is invoked or the automount(8) program
11 is run. Each line describes a mount point and refers to an autofs map
12 describing file systems to be mounted under the mount point.
13
14 The default location of the master map is /etc/auto.master but an al‐
15 ternate name may be given on the command line when running the auto‐
16 mounter and the default master map may changed by setting the
17 MASTER_MAP_NAME configuration variable in /etc/sysconfig/autofs. If
18 the master map name has no path then the system Name Service Switch
19 configuration will be consulted and each of the sources searched in
20 line with the rules given in the Name Service Switch configuration.
21
22 Access to mounts in maps is governed by a key.
23
24 For direct maps the mount point is always specified as:
25
26 /-
27
28 and the key used within the direct map is the full path to the mount
29 point. The direct map may have multiple entries in the master map.
30
31 For indirect maps access is by using the path scheme:
32
33 /mount-point/key
34
35 where mount-point is one of the entries listed in the master map. The
36 key is a single directory component and is matched against entries in
37 the map given in the entry (See autofs(5)).
38
39 Additionally, a map may be included from its source as if it were it‐
40 self present in the master map by including a line of the form:
41
42 +[maptype[,format]:]map [options]
43
44 and automount(8) will process the map according to the specification
45 described below for map entries. Plus map inclusion is only permitted
46 in file map sources. Indirect map entries must be unique in the master
47 map so second and subsequent entries for an indirect mount point are
48 ignored by automount(8).
49
50 NOTE: autofs currently does not collapse multiple slashes in paths, so
51 it is important to ensure paths used in maps are correct. If un‐
52 necessary multiple slashes are present in a path it can lead to
53 unexpected failures such as an inability to expire automounts.
54 An exception to this is a trailing slash at the end of the auto‐
55 mount point path in the master map which will be removed if
56 present.
57
59 Master map entries have three fields separated by an arbitrary number
60 of spaces or tabs. Lines beginning with # are comments. The first field
61 is the mount point described above and the second field is the name of
62 the map to be consulted for the mount point followed by the third field
63 which contains options to be applied to all entries in the map.
64
65 The format of a master map entry is:
66
67 mount-point [map-type[,format]:]map [options]
68
69 mount-point
70 Base location for the autofs filesystem to be mounted. For in‐
71 direct maps this directory will be created (as with mkdir -p)
72 and is removed when the autofs filesystem is umounted.
73
74 map-type
75 Type of map used for this mount point. The following are valid
76 map types:
77
78 file The map is a regular text file.
79
80 program
81 The map is an executable program, which is passed a key
82 on the command line and returns an entry (everything be‐
83 sides the key) on stdout if successful. Optinally, the
84 keyword exec may be used as a synonym for program to
85 avoid confusion with amd formated maps mount type pro‐
86 gram.
87
88 yp The map is a NIS (YP) database.
89
90 nisplus
91 The map is a NIS+ database.
92
93 hesiod The map is a hesiod database whose filsys entries are
94 used for maps.
95
96 ldap or ldaps
97 The map is stored in an LDAP directory. If ldaps is used
98 the appropriate certificate must be configured in the
99 LDAP client.
100
101 multi This map type allows the specification of multiple maps
102 separated by "--". These maps are searched in order to
103 resolve key lookups.
104
105 dir This map type can be used at + master map including nota‐
106 tion. The contents of files under given directory are in‐
107 cluded to the master map. The name of file to be included
108 must be ended with ".autofs". A file will be ignored if
109 its name is not ended with the suffix. In addition a dot
110 file, a file which name is started with "." is also ig‐
111 nored.
112
113 format
114 Format of the map data; currently the formats recognized are
115 sun, which is a subset of the Sun automounter map format, hes‐
116 iod, for hesiod filesys entries and amd for amd formated map en‐
117 tries. If the format is left unspecified, it defaults to sun
118 for all map types except hesiod unless it is a top level amd
119 mount that has a configuration entry for the mount point path,
120 in which case the format used is amd.
121
122 map
123 Name of the map to use. This is an absolute UNIX pathname for
124 maps of types file, dir, or program, and the name of a database
125 in the case for maps of type yp, nisplus, or hesiod or the dn of
126 an LDAP entry for maps of type ldap.
127
128 options
129 Any remaining command line arguments without leading dashes (-)
130 are taken as options (-o) to mount. Arguments with leading
131 dashes are considered options for the maps and are passed to au‐
132 tomount (8).
133
134 The sun format supports the following options:
135
136 -Dvariable=value
137 Replace variable with value in map substitutions.
138
139 -strict
140 Treat errors when mounting file systems as fatal. This is
141 important when multiple file systems should be mounted
142 (`multimounts'). If this option is given, no file system
143 is mounted at all if at least one file system can't be
144 mounted.
145
146 [no]browse
147 This is an autofs specific option that is a pseudo mount
148 option and so is given without a leading dash. Use of the
149 browse option pre-creates mount point directories for in‐
150 direct mount maps so the map keys can be seen in a direc‐
151 tory listing without being mounted. Use of this option
152 can cause performance problem if the indirect map is
153 large so it should be used with caution. The internal
154 program default is to enable browse mode for indirect
155 mounts but the default installed configuration overrides
156 this by setting BROWSE_MODE to "no" because of the poten‐
157 tial performance problem. This option does the same as
158 the deprecated --ghost option, the browse option is pre‐
159 ferred because it is used by other autofs implementa‐
160 tions.
161
162 nobind This is an autofs specific option that is a pseudo mount
163 option and so is given without a leading dash. It may be
164 used either in the master map entry (so it effects all
165 the map entries) or with individual map entries to pre‐
166 vent bind mounting of local NFS filesystems. For direct
167 mount maps the option is only effective if specified on
168 the first direct map entry and is applied to all direct
169 mount maps in the master map. It is ignored if given on
170 subsequent direct map entries. It may be used on individ‐
171 ual map entries of both types. Preventing bind mounts of
172 NFS file systems can no longer be done by using the
173 "port=" option, the nobind option must be used instead.
174
175 symlink
176 This option makes bind mounting use a symlink instead of
177 an actual bind mount. It is an autofs specific option
178 that is a pseudo mount option and so is given without a
179 leading dash. It may be used with indirect map entries
180 only, either in the master map (so it effects all map en‐
181 tries) or with individual map entries. The option is ig‐
182 nored for direct mounts and non-root offest mount en‐
183 tries.
184
185 strictexpire
186 Use a strict expire policy for this automount. Using this
187 option means that last use of autofs directory entries
188 will not be updated during path walks so that mounts in
189 an automount won't be kept mounted by applications scan‐
190 ning the mount tree. Note that this doesn't completely
191 resolve the problem of expired automounts being immedi‐
192 ately re-mounted due to application accesses triggered by
193 the expire itself.
194
195 slave, private or shared
196 This option allows mount propagation of bind mounts to be
197 set to slave, private or shared. This option defaults to
198 slave if no option is given. When using multi-mounts that
199 have bind mounts the bind mount will have the same prop‐
200 erties as its parent which is commonly propagation
201 shared. And if the mount target is also propagation
202 shared this can lead to a deadlock when attempting to ac‐
203 cess the offset mounts. When this happens an unwanted
204 offset mount is propagated back to the target file system
205 resulting in a deadlock since the automount target is it‐
206 self an (unwanted) automount trigger. This option is an
207 autofs pseudo mount option that can be used in the master
208 map only.
209
210 -r, --random-multimount-selection
211 Enables the use of random selection when choosing a host
212 from a list of replicated servers. This option is applied
213 to this mount only, overriding the global setting that
214 may be specified on the command line.
215
216 -w, --use-weight-only
217 Use only specified weights for server selection where
218 more than one server is specified in the map entry. If no
219 server weights are given then each available server will
220 be tried in the order listed, within proximity.
221
222 -t, --timeout <seconds>
223 Set the expire timeout for map entries. This option can
224 be used to override the global default given either on
225 the command line or in the configuration.
226
227 -n, --negative-timeout <seconds>
228 Set the timeout for caching failed key lookups. This op‐
229 tion can be used to override the global default given ei‐
230 ther on the command line or in the configuration.
231
232 --mode <octal_mode>
233 Set the directory mode for the base location of the aut‐
234 ofs mount point. If this option is given, autofs will
235 chmod that directory with this mode.
236
238 If "-hosts" is given as the map then accessing a key under the mount
239 point which corresponds to a hostname will allow access to the exports
240 of that host. The hosts map cannot be dynamically updated and requires
241 a HUP signal to be sent to the daemon for it to check hosts for an up‐
242 date. Due to possible hierarchic dependencies within a mount tree, it
243 might not be completely updated during the HUP signal processing.
244
245 For example, with an entry in the master map of /net -hosts accessing
246 /net/myserver will mount exports from myserver on directories below
247 /net/myserver.
248
249 NOTE: mounts done from a hosts map will be mounted with the "no‐
250 suid,nodev" options unless overridden by explicitly specifying the
251 "suid", "dev" options in the master map entry.
252
254 If the map type ldap is specified the mapname is of the form [//server‐
255 name/]dn, where the optional servername is the name of the LDAP server
256 to query, and dn is the Distinguished Name of a subtree to search for
257 map entries. The old style ldap:servername:mapname is also understood.
258 Alternatively, the type can be obtained from the Name Service Switch
259 configuration, in which case the map name alone must be given.
260
261 If no schema is set in the autofs configuration then autofs will check
262 each of the commonly used schema for a valid entry and if one is found
263 it will be used for subsequent lookups.
264
265 There are three common schemas in use:
266
267 nisMap
268 Entries in the nisMap schema are nisObject objects in the speci‐
269 fied subtree, where the cn attribute is the key (the wildcard
270 key is "/"), and the nisMapEntry attribute contains the informa‐
271 tion used by the automounter.
272
273 automountMap
274 The automountMap schema has two variations that differ in the
275 attribute used for the map key. Entries in the automountMap
276 schema are automount objects in the specified subtree, where the
277 cn or automountKey attribute (depending on local usage) is the
278 key (the wildcard key is "/"), and the automountInformation at‐
279 tribute contains the information used by the automounter. Note
280 that the cn attribute is case insensitive.
281
282 The object classes and attributes used for accessing automount maps in
283 LDAP can be changed by setting entries in the autofs configuration lo‐
284 cated in /etc/sysconfig/autofs.conf.
285
286 NOTE: If a schema is given in the configuration then all the schema
287 configuration values must be set, any partial schema specifica‐
288 tion will be ignored.
289
290 For amd format maps a different schema is used:
291
292 amdMap
293 The amdmap schema contains attributes amdmapName, amdmapKey and
294 amdmapValue where amdmapName contains the name of the containing
295 map, amdmapKey contains the map key and amdmapValue contains the
296 map entry.
297
299 LDAP authenticated binds, TLS encrypted connections and certification
300 may be used by setting appropriate values in the autofs authentication
301 configuration file and configuring the LDAP client with appropriate
302 settings. The default location of this file is
303 /etc/autofs_ldap_auth.conf.
304
305 If this file exists it will be used to establish whether TLS or authen‐
306 tication should be used.
307
308 An example of this file is:
309
310 <?xml version="1.0" ?>
311 <autofs_ldap_sasl_conf
312 usetls="yes"
313 tlsrequired="no"
314 authrequired="no"
315 authtype="DIGEST-MD5"
316 user="xyz"
317 secret="abc"
318 />
319
320 If TLS encryption is to be used the location of the Certificate Author‐
321 ity certificate must be set within the LDAP client configuration in or‐
322 der to validate the server certificate. If, in addition, a certified
323 connection is to be used then the client certificate and private key
324 file locations must also be configured within the LDAP client.
325
326 In OpenLDAP these may be configured in the ldap.conf file or in the
327 per-user configuration. For example, it may be sensible to use the sys‐
328 tem wide configuration for the location of the Certificate Authority
329 certificate and set the location of the client certificate and private
330 key in the per-user configuration. The location of these files and the
331 configuration entry requirements is system dependent so the documenta‐
332 tion for your installation will need to be consulted to get further in‐
333 formation.
334
335 See autofs_ldap_auth.conf (5) for more information.
336
338 /- auto.data
339 /home /etc/auto.home
340 /mnt yp:mnt.map
341
342 This will generate two mountpoints for /home and /mnt and install di‐
343 rect mount triggers for each entry in the direct mount map auto.data.
344 All accesses to /home will lead to the consultation of the map in
345 /etc/auto.home and all accesses to /mnt will consult the NIS map
346 mnt.map. All accesses to paths in the map auto.data will trigger
347 mounts when they are accessed and the Name Service Switch configuration
348 will be used to locate the source of the map auto.data.
349
350 To avoid making edits to /etc/auto.master, /etc/auto.master.d may be
351 used. Files in that directory must have a ".autofs" suffix, e.g.
352 /etc/auto.master.d/extra.autofs. Such files contain lines of the same
353 format as the auto.master file, e.g.
354
355 /foo /etc/auto.foo
356 /baz yp:baz.map
357
359 automount(8), autofs(5), autofs(8), autofs.conf(5), aut‐
360 ofs_ldap_auth.conf(5).
361
363 This manual page was written by Christoph Lameter <chris@waterf.org>,
364 for the Debian GNU/Linux system. Edited by <hpa@transmeta.com> and Ian
365 Kent <raven@themaw.net> .
366
367
368
369 11 Apr 2006 AUTO.MASTER(5)