xref: /freebsd/sbin/mount_nfs/mount_nfs.8 (revision a9e8641da961bcf3d24afc85fd657f2083a872a2)
1.\" Copyright (c) 1992, 1993, 1994, 1995
2.\"	The Regents of the University of California.  All rights reserved.
3.\"
4.\" Redistribution and use in source and binary forms, with or without
5.\" modification, are permitted provided that the following conditions
6.\" are met:
7.\" 1. Redistributions of source code must retain the above copyright
8.\"    notice, this list of conditions and the following disclaimer.
9.\" 2. Redistributions in binary form must reproduce the above copyright
10.\"    notice, this list of conditions and the following disclaimer in the
11.\"    documentation and/or other materials provided with the distribution.
12.\" 4. Neither the name of the University nor the names of its contributors
13.\"    may be used to endorse or promote products derived from this software
14.\"    without specific prior written permission.
15.\"
16.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
17.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
18.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
19.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
20.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
21.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
22.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
23.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
24.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
25.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
26.\" SUCH DAMAGE.
27.\"
28.\"	@(#)mount_nfs.8	8.3 (Berkeley) 3/29/95
29.\" $FreeBSD$
30.\"
31.Dd December 7, 2013
32.Dt MOUNT_NFS 8
33.Os
34.Sh NAME
35.Nm mount_nfs
36.Nd mount NFS file systems
37.Sh SYNOPSIS
38.Nm
39.Op Fl 23bcdiLlNPsTU
40.Op Fl a Ar maxreadahead
41.Op Fl D Ar deadthresh
42.Op Fl g Ar maxgroups
43.Op Fl I Ar readdirsize
44.Op Fl o Ar options
45.Op Fl R Ar retrycnt
46.Op Fl r Ar readsize
47.Op Fl t Ar timeout
48.Op Fl w Ar writesize
49.Op Fl x Ar retrans
50.Ar rhost : Ns Ar path node
51.Sh DESCRIPTION
52The
53.Nm
54utility calls the
55.Xr nmount 2
56system call to prepare and graft a remote NFS file system
57.Pq Ar rhost : Ns Ar path
58on to the file system tree at the point
59.Ar node .
60This command is normally executed by
61.Xr mount 8 .
62It implements the mount protocol as described in RFC 1094, Appendix A and
63.%T "NFS: Network File System Version 3 Protocol Specification" ,
64Appendix I.
65.Pp
66If the file system type is specified as ``oldnfs'', which implies this
67command is run as ``mount_oldnfs'', then it forces use of the old NFS
68client, which does not support the
69.Cm nfsv4
70option.
71.Pp
72By default,
73.Nm
74keeps retrying until the mount succeeds.
75This behaviour is intended for file systems listed in
76.Xr fstab 5
77that are critical to the boot process.
78For non-critical file systems, the
79.Cm bg
80and
81.Cm retrycnt
82options provide mechanisms to prevent the boot process from hanging
83if the server is unavailable.
84.Pp
85If the server becomes unresponsive while an NFS file system is
86mounted, any new or outstanding file operations on that file system
87will hang uninterruptibly until the server comes back.
88To modify this default behaviour, see the
89.Cm intr
90and
91.Cm soft
92options.
93.Pp
94The options are:
95.Bl -tag -width indent
96.It Fl o
97Options are specified with a
98.Fl o
99flag followed by a comma separated string of options.
100See the
101.Xr mount 8
102man page for possible options and their meanings.
103The following NFS specific options are also available:
104.Bl -tag -width indent
105.It Cm acregmin Ns = Ns Aq Ar seconds
106.It Cm acregmax Ns = Ns Aq Ar seconds
107.It Cm acdirmin Ns = Ns Aq Ar seconds
108.It Cm acdirmax Ns = Ns Aq Ar seconds
109When attributes of files are cached, a timeout calculated to determine
110whether a given cache entry has expired.
111These four values determine the upper and lower bounds of the timeouts for
112.Dq directory
113attributes and
114.Dq regular
115(ie: everything else).
116The default values are 3 -> 60 seconds
117for regular files, and 30 -> 60 seconds for directories.
118The algorithm to calculate the timeout is based on the age of the file.
119The older the file,
120the longer the cache is considered valid, subject to the limits above.
121.It Cm allgssname
122This option can be used along with
123.Fl o Cm gssname
124to specify that all operations should use the host-based initiator
125credential.
126This may be used for clients that run system daemons that need to
127access files on the NFSv4 mounted volume.
128.It Cm bg
129If an initial attempt to contact the server fails, fork off a child to keep
130trying the mount in the background.
131Useful for
132.Xr fstab 5 ,
133where the file system mount is not critical to multiuser operation.
134.It Cm deadthresh Ns = Ns Aq Ar value
135Set the
136.Dq "dead server threshold"
137to the specified number of round trip timeout intervals before a
138.Dq "server not responding"
139message is displayed.
140.It Cm dumbtimer
141Turn off the dynamic retransmit timeout estimator.
142This may be useful for UDP mounts that exhibit high retry rates,
143since it is possible that the dynamically estimated timeout interval is too
144short.
145.It Cm fg
146Same as not specifying
147.Cm bg .
148.It Cm gssname Ns = Ns Aq Ar service-principal-name
149This option can be used with the KerberosV security flavors for NFSv4 mounts
150to specify the
151.Dq "service-principal-name"
152of a host-based entry in the default
153keytab file that is used for system operations.
154It allows the mount to be performed by
155.Dq "root"
156and avoids problems with
157cached credentials for the system operations expiring.
158The
159.Dq "service-prinicpal-name"
160should be specified without instance or domain and is typically
161.Dq "host" ,
162.Dq "nfs"
163or
164.Dq "root" .
165.It Cm hard
166Same as not specifying
167.Cm soft .
168.It Cm intr
169Make the mount interruptible, which implies that file system calls that
170are delayed due to an unresponsive server will fail with EINTR when a
171termination signal is posted for the process.
172.It Cm maxgroups Ns = Ns Aq Ar value
173Set the maximum size of the group list for the credentials to the
174specified value.
175This should be used for mounts on old servers that cannot handle a
176group list size of 16, as specified in RFC 1057.
177Try 8, if users in a lot of groups cannot get response from the mount
178point.
179.It Cm mntudp
180Force the mount protocol to use UDP transport, even for TCP NFS mounts.
181(Necessary for some old
182.Bx
183servers.)
184.It Cm nametimeo Ns = Ns Aq Ar value
185Override the default of NFS_DEFAULT_NAMETIMEO for the timeout (in seconds)
186for positive name cache entries.
187If this is set to 0 it disables positive name caching for the mount point.
188.It Cm negnametimeo Ns = Ns Aq Ar value
189Override the default of NFS_DEFAULT_NEGNAMETIMEO for the timeout (in seconds)
190for negative name cache entries. If this is set to 0 it disables negative
191name caching for the mount point.
192.It Cm nfsv2
193Use the NFS Version 2 protocol (the default is to try version 3 first
194then version 2).
195Note that NFS version 2 has a file size limit of 2 gigabytes.
196.It Cm nfsv3
197Use the NFS Version 3 protocol.
198.It Cm nfsv4
199Use the NFS Version 4 protocol.
200This option will force the mount to use
201TCP transport.
202.It Cm minorversion Ns = Ns Aq Ar value
203Override the default of 0 for the minor version of the NFS Version 4 protocol.
204The only minor version currently supported is 1.
205This option is only meaningful when used with the
206.Cm nfsv4
207option.
208.It Cm pnfs
209Enable support for parallel NFS (pNFS) for minor version 1 of the
210NFS Version 4 protocol.
211This option is only meaningful when used with the
212.Cm minorversion
213option.
214.It Cm noconn
215For UDP mount points, do not do a
216.Xr connect 2 .
217This must be used if the server does not reply to requests from the standard
218NFS port number 2049 or replies to requests using a different IP address
219(which can occur if the server is multi-homed).
220Setting the
221.Va vfs.nfs.nfs_ip_paranoia
222sysctl to 0 will make this option the default.
223.It Cm nocto
224Normally, NFS clients maintain the close-to-open cache coherency.
225This works by flushing at close time and checking at open time.
226Checking at open time is implemented by getting attributes from
227the server and purging the data cache if they do not match
228attributes cached by the client.
229.Pp
230This option disables checking at open time.
231It may improve performance for read-only mounts,
232but should only be used if the data on the server changes rarely.
233Be sure to understand the consequences before enabling this option.
234.It Cm noinet4 , noinet6
235Disables
236.Dv AF_INET
237or
238.Dv AF_INET6
239connections.
240Useful for hosts that have
241both an A record and an AAAA record for the same name.
242.It Cm nolockd
243Do
244.Em not
245forward
246.Xr fcntl 2
247locks over the wire.
248All locks will be local and not seen by the server
249and likewise not seen by other NFS clients.
250This removes the need to run the
251.Xr rpcbind 8
252service and the
253.Xr rpc.statd 8
254and
255.Xr rpc.lockd 8
256servers on the client.
257Note that this option will only be honored when performing the
258initial mount, it will be silently ignored if used while updating
259the mount options.
260.It Cm noncontigwr
261This mount option allows the NFS client to
262combine non-contiguous byte ranges being written
263such that the dirty byte range becomes a superset of the bytes
264that are dirty.
265This reduces the number of writes significantly for software
266builds.
267The merging of byte ranges isn't done if the file has been file
268locked, since most applications modifying a file from multiple
269clients will use file locking.
270As such, this option could result in a corrupted file for the
271rare case of an application modifying the file from multiple
272clients concurrently without using file locking.
273.It Cm principal
274For the RPCSEC_GSS security flavors, such as krb5, krb5i and krb5p,
275this option sets the name of the host based principal name expected
276by the server. This option overrides the default, which will be
277``nfs@<server-fqdn>'' and should normally be sufficient.
278.It Cm noresvport
279Do
280.Em not
281use a reserved socket port number (see below).
282.It Cm port Ns = Ns Aq Ar port_number
283Use specified port number for NFS requests.
284The default is to query the portmapper for the NFS port.
285.It Cm rdirplus
286Used with NFSV3 to specify that the \fBReaddirPlus\fR RPC should
287be used.
288For NFSV4, setting this option has a similar effect, in that it will make
289the Readdir Operation get more attributes.
290This option reduces RPC traffic for cases such as
291.Dq "ls -l" ,
292but tends to flood the attribute and name caches with prefetched entries.
293Try this option and see whether performance improves or degrades.
294Probably
295most useful for client to server network interconnects with a large bandwidth
296times delay product.
297.It Cm readahead Ns = Ns Aq Ar value
298Set the read-ahead count to the specified value.
299This may be in the range of 0 - 4, and determines how many blocks
300will be read ahead when a large file is being read sequentially.
301Trying a value greater than 1 for this is suggested for
302mounts with a large bandwidth * delay product.
303.It Cm readdirsize Ns = Ns Aq Ar value
304Set the readdir read size to the specified value.
305The value should normally
306be a multiple of
307.Dv DIRBLKSIZ
308that is <= the read size for the mount.
309.It Cm resvport
310Use a reserved socket port number.
311This flag is obsolete, and only retained for compatibility reasons.
312Reserved port numbers are used by default now.
313(For the rare case where the client has a trusted root account
314but untrustworthy users and the network cables are in secure areas this does
315help, but for normal desktop clients this does not apply.)
316.It Cm retrans Ns = Ns Aq Ar value
317Set the retransmit timeout count for soft mounts to the specified value.
318.It Cm retrycnt Ns = Ns Aq Ar count
319Set the mount retry count to the specified value.
320The default is a retry count of zero, which means to keep retrying
321forever.
322There is a 60 second delay between each attempt.
323.It Cm rsize Ns = Ns Aq Ar value
324Set the read data size to the specified value.
325It should normally be a power of 2 greater than or equal to 1024.
326This should be used for UDP mounts when the
327.Dq "fragments dropped due to timeout"
328value is getting large while actively using a mount point.
329(Use
330.Xr netstat 1
331with the
332.Fl s
333option to see what the
334.Dq "fragments dropped due to timeout"
335value is.)
336.It Cm sec Ns = Ns Aq Ar flavor
337This option specifies what security flavor should be used for the mount.
338Currently, they are:
339.Bd -literal
340krb5 -  Use KerberosV authentication
341krb5i - Use KerberosV authentication and
342        apply integrity checksums to RPCs
343krb5p - Use KerberosV authentication and
344        encrypt the RPC data
345sys -   The default AUTH_SYS, which uses a
346        uid + gid list authenticator
347.Ed
348.It Cm soft
349A soft mount, which implies that file system calls will fail
350after
351.Ar retrycnt
352round trip timeout intervals.
353.It Cm tcp
354Use TCP transport.
355This is the default option, as it provides for increased reliability on both
356LAN and WAN configurations compared to UDP.
357Some old NFS servers do not support this method; UDP mounts may be required
358for interoperability.
359.It Cm timeout Ns = Ns Aq Ar value
360Set the initial retransmit timeout to the specified value.
361May be useful for fine tuning UDP mounts over internetworks
362with high packet loss rates or an overloaded server.
363Try increasing the interval if
364.Xr nfsstat 1
365shows high retransmit rates while the file system is active or reducing the
366value if there is a low retransmit rate but long response delay observed.
367(Normally, the
368.Cm dumbtimer
369option should be specified when using this option to manually
370tune the timeout
371interval.)
372.It Cm udp
373Use UDP transport.
374.It Cm wcommitsize Ns = Ns Aq Ar value
375Set the maximum pending write commit size to the specified value.
376This determines the maximum amount of pending write data that the NFS
377client is willing to cache for each file.
378.It Cm wsize Ns = Ns Aq Ar value
379Set the write data size to the specified value.
380Ditto the comments w.r.t.\& the
381.Cm rsize
382option, but using the
383.Dq "fragments dropped due to timeout"
384value on the server instead of the client.
385Note that both the
386.Cm rsize
387and
388.Cm wsize
389options should only be used as a last ditch effort at improving performance
390when mounting servers that do not support TCP mounts.
391.El
392.El
393.Sh COMPATIBILITY
394The following command line flags are equivalent to
395.Fl o
396named options and are supported for compatibility with older
397installations.
398.Bl -tag -width indent
399.It Fl 2
400Same as
401.Fl o Cm nfsv2
402.It Fl 3
403Same as
404.Fl o Cm nfsv3
405.It Fl D
406Same as
407.Fl o Cm deadthresh
408.It Fl I
409Same as
410.Fl o Cm readdirsize Ns = Ns Aq Ar value
411.It Fl L
412Same as
413.Fl o Cm nolockd
414.It Fl N
415Same as
416.Fl o Cm noresvport
417.It Fl P
418Use a reserved socket port number.
419This flag is obsolete, and only retained for compatibility reasons.
420(For the rare case where the client has a trusted root account
421but untrustworthy users and the network cables are in secure areas this does
422help, but for normal desktop clients this does not apply.)
423.It Fl R
424Same as
425.Fl o Cm retrycnt Ns = Ns Aq Ar value
426.It Fl T
427Same as
428.Fl o Cm tcp
429.It Fl U
430Same as
431.Fl o Cm mntudp
432.It Fl a
433Same as
434.Fl o Cm readahead Ns = Ns Aq Ar value
435.It Fl b
436Same as
437.Fl o Cm bg
438.It Fl c
439Same as
440.Fl o Cm noconn
441.It Fl d
442Same as
443.Fl o Cm dumbtimer
444.It Fl g
445Same as
446.Fl o Cm maxgroups
447.It Fl i
448Same as
449.Fl o Cm intr
450.It Fl l
451Same as
452.Fl o Cm rdirplus
453.It Fl r
454Same as
455.Fl o Cm rsize Ns = Ns Aq Ar value
456.It Fl s
457Same as
458.Fl o Cm soft
459.It Fl t
460Same as
461.Fl o Cm retransmit Ns = Ns Aq Ar value
462.It Fl w
463Same as
464.Fl o Cm wsize Ns = Ns Aq Ar value
465.It Fl x
466Same as
467.Fl o Cm retrans Ns = Ns Aq Ar value
468.El
469.Sh SEE ALSO
470.Xr nmount 2 ,
471.Xr unmount 2 ,
472.Xr nfsv4 4 ,
473.Xr fstab 5 ,
474.Xr gssd 8 ,
475.Xr mount 8 ,
476.Xr nfsd 8 ,
477.Xr nfsiod 8 ,
478.Xr showmount 8
479.Sh BUGS
480Since nfsv4 performs open/lock operations that have their ordering strictly
481enforced by the server, the options
482.Cm intr
483and
484.Cm soft
485cannot be safely used.
486.Cm hard
487nfsv4 mounts are strongly recommended.
488