1 --- /dev/null 2006-10-11 11:03:58.566320005 -0400
2 +++ nfs-utils-1.0.9/utils/mount/nfs.man 2006-10-16 08:31:02.204578000 -0400
4 +.\" nfs.5 "Rick Sladkey" <jrs@world.std.com>
5 +.\" Wed Feb 8 12:52:42 1995, faith@cs.unc.edu: updates for Ross Biro's
7 +.TH NFS 5 "20 November 1993" "Linux 0.99" "Linux Programmer's Manual"
9 +nfs \- nfs and nfs4 fstab format and options
15 +file contains information about which filesystems
16 +to mount where and with what options.
17 +For NFS mounts, it contains the server name and
18 +exported server directory to mount from,
19 +the local directory that is the mount point,
20 +and the NFS specific options that control
21 +the way the filesystem is mounted.
23 +Three different versions of the NFS protocol are
24 +supported by the Linux NFS client:
25 +NFS version 2, NFS version 3, and NFS version 4.
26 +To mount via NFS version 2, use the
28 +file system type and specify
30 +Version 2 is the default protocol version for the
32 +file system type when
34 +is not specified on the mount command.
35 +To mount via NFS version 3, use the
37 +file system type and specify
39 +To mount via NFS version 4, use the
44 +keyword is not supported for the
48 +These file system types share similar mount options;
49 +the differences are listed below.
51 +Here is an example from an \fI/etc/fstab\fP file for an NFSv2 mount
55 +.ta 2.5i +0.75i +0.75i +1.0i
56 +server:/usr/local/pub /pub nfs rsize=32768,wsize=32768,timeo=14,intr
59 +Here is an example for an NFSv4 mount over TCP using Kerberos
60 +5 mutual authentication.
63 +.ta 2.5i +0.75i +0.75i +1.0i
64 +server:/usr/local/pub /pub nfs4 proto=tcp,sec=krb5,hard,intr
67 +.SS Options for the nfs file system type
70 +The number of bytes NFS uses when reading files from an NFS server.
71 +The rsize is negotiated between the server and client to determine
72 +the largest block size that both can support.
73 +The value specified by this option is the maximum size that could
74 +be used; however, the actual size used may be smaller.
75 +Note: Setting this size to a value less than the largest supported
76 +block size will adversely affect performance.
79 +The number of bytes NFS uses when writing files to an NFS server.
80 +The wsize is negotiated between the server and client to determine
81 +the largest block size that both can support.
82 +The value specified by this option is the maximum size that could
83 +be used; however, the actual size used may be smaller.
84 +Note: Setting this size to a value less than the largest supported
85 +block size will adversely affect performance.
88 +The value in tenths of a second before sending the
89 +first retransmission after an RPC timeout.
90 +The default value is 7 tenths of a second. After the first timeout,
91 +the timeout is doubled after each successive timeout until a maximum
92 +timeout of 60 seconds is reached or the enough retransmissions
93 +have occured to cause a major timeout. Then, if the filesystem
94 +is hard mounted, each new timeout cascade restarts at twice the
95 +initial value of the previous cascade, again doubling at each
96 +retransmission. The maximum timeout is always 60 seconds.
97 +Better overall performance may be achieved by increasing the
98 +timeout when mounting on a busy network, to a slow server, or through
99 +several routers or gateways.
102 +The number of minor timeouts and retransmissions that must occur before
103 +a major timeout occurs. The default is 3 timeouts. When a major timeout
104 +occurs, the file operation is either aborted or a "server not responding"
105 +message is printed on the console.
108 +The minimum time in seconds that attributes of a regular file should
109 +be cached before requesting fresh information from a server.
110 +The default is 3 seconds.
113 +The maximum time in seconds that attributes of a regular file can
114 +be cached before requesting fresh information from a server.
115 +The default is 60 seconds.
118 +The minimum time in seconds that attributes of a directory should
119 +be cached before requesting fresh information from a server.
120 +The default is 30 seconds.
123 +The maximum time in seconds that attributes of a directory can
124 +be cached before requesting fresh information from a server.
125 +The default is 60 seconds.
128 +Using actimeo sets all of
135 +There is no default value.
138 +The number of minutes to retry an NFS mount operation
139 +in the foreground or background before giving up.
140 +The default value for forground mounts is 2 minutes.
141 +The default value for background mounts is 10000 minutes,
142 +which is roughly one week.
145 +When an NFS server does not support version two of the
146 +RPC mount protocol, this option can be used to specify
147 +the maximum length of a filename that is supported on
148 +the remote filesystem. This is used to support the
149 +POSIX pathconf functions. The default is 255 characters.
152 +The numeric value of the port to connect to the NFS server on.
153 +If the port number is 0 (the default) then query the
154 +remote host's portmapper for the port number to use.
155 +If the remote host's NFS daemon is not registered with
156 +its portmapper, the standard NFS port number 2049 is
160 +The numeric value of the
165 +The name of the host running
169 +Use an alternate RPC program number to contact the
170 +mount daemon on the remote host. This option is useful
171 +for hosts that can run multiple NFS servers.
172 +The default value is 100005 which is the standard RPC
173 +mount daemon program number.
176 +Use an alternate RPC version number to contact the
177 +mount daemon on the remote host. This option is useful
178 +for hosts that can run multiple NFS servers.
179 +The default value depends on which kernel you are using.
182 +Use an alternate RPC program number to contact the
183 +NFS daemon on the remote host. This option is useful
184 +for hosts that can run multiple NFS servers.
185 +The default value is 100003 which is the standard RPC
186 +NFS daemon program number.
189 +Use an alternate RPC version number to contact the
190 +NFS daemon on the remote host. This option is useful
191 +for hosts that can run multiple NFS servers.
192 +The default value depends on which kernel you are using.
195 +vers is an alternative to nfsvers and is compatible with
196 +many other operating systems.
199 +Disable NFS locking. Do not start lockd.
200 +This has to be used with some old NFS servers
201 +that don't support locking.
204 +If the first NFS mount attempt times out, retry the mount
206 +After a mount operation is backgrounded, all subsequent mounts
207 +on the same NFS server will be backgrounded immediately, without
208 +first attempting the mount.
209 +A missing mount point is treated as a timeout,
210 +to allow for nested NFS mounts.
213 +If the first NFS mount attempt times out, retry the mount
215 +This is the complement of the
217 +option, and also the default behavior.
220 +If an NFS file operation has a major timeout then report an I/O error to
221 +the calling program.
222 +The default is to continue retrying NFS file operations indefinitely.
225 +If an NFS file operation has a major timeout then report
226 +"server not responding" on the console and continue retrying indefinitely.
227 +This is the default.
230 +If an NFS file operation has a major timeout and it is hard mounted,
231 +then allow signals to interupt the file operation and cause it to
232 +return EINTR to the calling program. The default is to not
233 +allow file operations to be interrupted.
236 +Mount the NFS filesystem using POSIX semantics. This allows
237 +an NFS filesystem to properly support the POSIX pathconf
238 +command by querying the mount server for the maximum length
239 +of a filename. To do this, the remote host must support version
240 +two of the RPC mount protocol. Many NFS servers support only
244 +Suppress the retrieval of new attributes when creating a file.
247 +Disable all forms of attribute caching entirely. This extracts a
248 +significant performance penalty but it allows two different NFS clients
249 +to get reasonable results when both clients are actively
250 +writing to a common export on the server.
253 +Disables Access Control List (ACL) processing.
256 +Set the security flavor for this mount to "mode".
257 +The default setting is \f3sec=sys\f1, which uses local
258 +unix uids and gids to authenticate NFS operations (AUTH_SYS).
259 +Other currently supported settings are:
260 +\f3sec=krb5\f1, which uses Kerberos V5 instead of local unix uids
261 +and gids to authenticate users;
262 +\f3sec=krb5i\f1, which uses Kerberos V5 for user authentication
263 +and performs integrity checking of NFS operations using secure
264 +checksums to prevent data tampering; and
265 +\f3sec=krb5p\f1, which uses Kerberos V5 for user authentication
266 +and integrity checking, and encrypts NFS traffic to prevent
267 +traffic sniffing (this is the most secure setting).
268 +Note that there is a performance penalty when using integrity
272 +Mount the NFS filesystem using the TCP protocol instead of the
273 +default UDP protocol. Many NFS servers only support UDP.
276 +Mount the NFS filesystem using the UDP protocol. This
279 +All of the non-value options have corresponding nooption forms.
280 +For example, nointr means don't allow file operations to be
282 +.SS Options for the nfs4 file system type
285 +The number of bytes nfs4 uses when reading files from the server.
286 +The rsize is negotiated between the server and client to determine
287 +the largest block size that both can support.
288 +The value specified by this option is the maximum size that could
289 +be used; however, the actual size used may be smaller.
290 +Note: Setting this size to a value less than the largest supported
291 +block size will adversely affect performance.
294 +The number of bytes nfs4 uses when writing files to the server.
295 +The wsize is negotiated between the server and client to determine
296 +the largest block size that both can support.
297 +The value specified by this option is the maximum size that could
298 +be used; however, the actual size used may be smaller.
299 +Note: Setting this size to a value less than the largest supported
300 +block size will adversely affect performance.
303 +The value in tenths of a second before sending the
304 +first retransmission after an RPC timeout.
305 +The default value depends on whether
309 +is in effect (see below).
310 +The default value for UDP is 7 tenths of a second.
311 +The default value for TCP is 60 seconds.
312 +After the first timeout,
313 +the timeout is doubled after each successive timeout until a maximum
314 +timeout of 60 seconds is reached or the enough retransmissions
315 +have occured to cause a major timeout. Then, if the filesystem
316 +is hard mounted, each new timeout cascade restarts at twice the
317 +initial value of the previous cascade, again doubling at each
318 +retransmission. The maximum timeout is always 60 seconds.
321 +The number of minor timeouts and retransmissions that must occur before
322 +a major timeout occurs. The default is 5 timeouts for
326 +When a major timeout
327 +occurs, the file operation is either aborted or a "server not responding"
328 +message is printed on the console.
331 +The minimum time in seconds that attributes of a regular file should
332 +be cached before requesting fresh information from a server.
333 +The default is 3 seconds.
336 +The maximum time in seconds that attributes of a regular file can
337 +be cached before requesting fresh information from a server.
338 +The default is 60 seconds.
341 +The minimum time in seconds that attributes of a directory should
342 +be cached before requesting fresh information from a server.
343 +The default is 30 seconds.
346 +The maximum time in seconds that attributes of a directory can
347 +be cached before requesting fresh information from a server.
348 +The default is 60 seconds.
351 +Using actimeo sets all of
358 +There is no default value.
361 +The number of minutes to retry an NFS mount operation
362 +in the foreground or background before giving up.
363 +The default value for forground mounts is 2 minutes.
364 +The default value for background mounts is 10000 minutes,
365 +which is roughly one week.
368 +The numeric value of the port to connect to the NFS server on.
369 +If the port number is 0 (the default) then query the
370 +remote host's portmapper for the port number to use.
371 +If the remote host's NFS daemon is not registered with
372 +its portmapper, the standard NFS port number 2049 is
376 +Mount the NFS filesystem using a specific network protocol
377 +instead of the default UDP protocol.
378 +Many NFS version 4 servers only support TCP.
379 +Valid protocol types are
385 +On a multi-homed client, this
386 +causes the client to use a specific callback address when
387 +communicating with an NFS version 4 server.
388 +This option is currently ignored.
391 +Same as \f3sec=mode\f1 for the nfs filesystem type (see above).
394 +If an NFS mount attempt times out, retry the mount
396 +After a mount operation is backgrounded, all subsequent mounts
397 +on the same NFS server will be backgrounded immediately, without
398 +first attempting the mount.
399 +A missing mount point is treated as a timeout,
400 +to allow for nested NFS mounts.
403 +If the first NFS mount attempt times out, retry the mount
405 +This is the complement of the
407 +option, and also the default behavior.
410 +If an NFS file operation has a major timeout then report an I/O error to
411 +the calling program.
412 +The default is to continue retrying NFS file operations indefinitely.
415 +If an NFS file operation has a major timeout then report
416 +"server not responding" on the console and continue retrying indefinitely.
417 +This is the default.
420 +If an NFS file operation has a major timeout and it is hard mounted,
421 +then allow signals to interupt the file operation and cause it to
422 +return EINTR to the calling program. The default is to not
423 +allow file operations to be interrupted.
426 +Suppress the retrieval of new attributes when creating a file.
429 +Disable attribute caching, and force synchronous writes.
431 +server performance penalty but it allows two different NFS clients
432 +to get reasonable good results when both clients are actively
433 +writing to common filesystem on the server.
435 +All of the non-value options have corresponding nooption forms.
436 +For example, nointr means don't allow file operations to be
441 +.BR fstab "(5), " mount "(8), " umount "(8), " exports (5)
443 +"Rick Sladkey" <jrs@world.std.com>
445 +The posix, and nocto options are parsed by mount
446 +but currently are silently ignored.
448 +The tcp and namlen options are implemented but are not currently
449 +supported by the Linux kernel.
451 +The umount command should notify the server
452 +when an NFS filesystem is unmounted.
454 +Checking files on NFS filesystem referenced by file descriptors (i.e. the
458 +families of functions) may lead to inconsistent result due to the lack of
459 +consistency check in kernel even if noac is used.
460 --- nfs-utils-1.0.9/utils/mount/Makefile.am.orig 2006-10-16 08:29:34.790243000 -0400
461 +++ nfs-utils-1.0.9/utils/mount/Makefile.am 2006-10-16 08:30:32.594308000 -0400
463 ## Process this file with automake to produce Makefile.in
465 man8_MANS = mount.nfs.man umount.nfs.man
468 sbin_PROGRAMS = mount.nfs
469 -EXTRA_DIST = nfsmount.x $(man8_MANS)
470 +EXTRA_DIST = nfsmount.x $(man8_MANS) $(man5_MANS)
471 mount_nfs_SOURCES = mount.c nfsmount.c nfs4mount.c nfsumount.c \
472 mount_constants.h nfs4_mount.h nfs_mount4.h
474 @@ -29,6 +30,11 @@ install-man-links:
475 inst=`echo $$m | sed -e 's/man$$/8/'`; \
478 + (cd $(DESTDIR)$(man5dir) && \
479 + for m in $(man5_MANS) $(dist_man5_MANS) $(nodist_man5_MANS); do \
480 + inst=`echo $$m | sed -e 's/man$$/5/'`; \
485 (cd $(DESTDIR)$(man8dir) && \
486 @@ -36,4 +42,9 @@ uninstall-man-links:
487 inst=`echo $$m | sed -e 's/man$$/8/'`; \
490 + (cd $(DESTDIR)$(man5dir) && \
491 + for m in $(man5_MANS) $(dist_man5_MANS) $(nodist_man5_MANS); do \
492 + inst=`echo $$m | sed -e 's/man$$/5/'`; \