1.\" Copyright (c) 1992, 1993, 1994 2.\" The Regents of the University of California. All rights reserved. 3.\" 4.\" %sccs.include.redist.roff% 5.\" 6.\" @(#)mount_nfs.8 8.2 (Berkeley) 03/27/94 7.\" 8.Dd 9.Dt MOUNT_NFS 8 10.Os BSD 4.4 11.Sh NAME 12.Nm mount_nfs 13.Nd mount nfs file systems 14.Sh SYNOPSIS 15.Nm mount_nfs 16.Op Fl KMPTbcdiklqs 17.Op Fl D Ar deadthresh 18.Op Fl L Ar leaseterm 19.Op Fl R Ar retrycnt 20.Op Fl a Ar maxreadahead 21.Op Fl g Ar maxgroups 22.Op Fl m Ar realm 23.Op Fl o Ar options 24.Op Fl r Ar readsize 25.Op Fl t Ar timeout 26.Op Fl w Ar writesize 27.Op Fl x Ar retrans 28.Ar rhost:path node 29.Sh DESCRIPTION 30The 31.Nm mount_nfs 32command 33calls the 34.Xr mount 2 35system call to prepare and graft a remote nfs file system (rhost:path) 36on to the file system tree at the point 37.Ar node. 38This command is normally executed by 39.Xr mount 8 . 40It implements the mount protocol as described in RFC 1094, Appendix A. 41.Pp 42The options are: 43.Bl -tag -width indent 44.It Fl D 45Used with NQNFS to set the 46.Dq "dead server threshold" 47to the specified number of round trip timeout intervals. 48After a 49.Dq "dead server threshold" 50of retransmit timeouts, 51cached data for the unresponsive server is assumed to still be valid. 52Values may be set in the range of 1 - 9, with 9 referring to an 53.Dq "infinite dead threshold" 54(i.e. never assume cached data still valid). 55This option is not generally recommended and is really an experimental 56feature. 57.It Fl K 58Pass Kerberos authenticators to the server for client-to-server 59user-credential mapping. 60This may only be used over TCP mounts between 4.4BSD clients and servers. 61.It Fl L 62Used with NQNFS to set the lease term to the specified number of seconds. 63Only use this argument for mounts with a large round trip delay. 64Values are normally in the 10-30 second range. 65.It Fl M 66Assume that other clients are not writing a file concurrently with this client. 67This implements a slightly less strict consistency criteria than 4.3BSD-Reno 68did, that is more in line with most commercial client implementations. 69This is recommended for servers that do not support leasing. 70.It Fl P 71Use a reserved socket port number. 72This is useful for mounting servers that require clients to use a 73reserved port number. 74.It Fl R 75Set the retry count for doing the mount to the specified value. 76.It Fl T 77Use TCP transport instead of UDP. 78This is recommended for servers that are not on the same LAN cable as 79the client. 80(NB: This is NOT supported by most non-BSD servers.) 81.It Fl a 82Set the read-ahead count to the specified value. 83This may be in the range of 0 - 4, and determines how many blocks 84will be read ahead when a large file is being read sequentially. 85This is recommended for mounts with a large bandwidth * delay product. 86.It Fl b 87If an initial attempt to contact the server fails, fork off a child to keep 88trying the mount in the background. 89Useful for 90.Xr fstab 5 , 91where the filesystem mount is not critical to multiuser operation. 92.It Fl c 93For UDP mount points, do not do a 94.Xr connect 2 . 95This must be used for servers that do not reply to requests from the 96standard port number. 97.It Fl d 98Do not estimate retransmit timeout dynamically. 99This may be useful for UDP mounts that exhibit high retry rates. 100.It Fl g 101Set the maximum size of the group list for the credentials to the 102specified value. 103This should be used for mounts on old servers that cannot handle a 104group list size of 16, as specified in RFC 1057. 105Try 8, if users in a lot of groups cannot get response from the mount 106point. 107.It Fl i 108Make the mount interruptible, which implies that file system calls that 109are delayed due to an unresponsive server will fail with EINTR when a 110termination signal is posted for the process. 111.It Fl k 112Used with NQNFS to specify 113.Dq get a lease 114for the file name being looked up. 115This is recommended unless the server is complaining about excessive 116lease load. 117.It Fl l 118Used with NQNFS to specify that the \fBReaddir_and_Lookup\fR RPC should 119be used. 120This option reduces RPC traffic for cases such as 121.Dq "ls -l" , 122but increases the lease load on the server. 123This is recommended unless the server is complaining about excessive 124lease load. 125.It Fl m 126Set the Kerberos realm to the string argument. 127Used with the 128.Fl K 129option for mounts to other realms. 130.It Fl o 131Options are specified with a 132.Fl o 133flag followed by a comma separated string of options. 134See the 135.Xr mount 8 136man page for possible options and their meanings. 137.It Fl q 138Use the leasing extensions to the protocol to maintain cache consistency. 139This protocol, referred to as Not Quite Nfs (NQNFS), 140is only supported by 4.4BSD servers. 141.It Fl r 142Set the read data size to the specified value. 143It should be a power of 2 greater than or equal to 1024. 144This should be used for UDP mounts when the 145.Dq "fragments dropped due to timeout" 146value is getting large while actively using a mount point. 147(Use 148.Xr netstat 1 149with the 150.Fl s 151option to see what the 152.Dq "fragments dropped due to timeout" 153value is.) 154See the 155.Fl w 156option as well. 157.It Fl s 158A soft mount, which implies that file system calls will fail 159after \fBRetry\fR round trip timeout intervals. 160.It Fl t 161Set the initial retransmit timeout to the specified value. 162May be useful for fine tuning UDP mounts over internetworks 163with high packet loss rates or an overloaded server. 164Try increasing the interval if 165.Xr nfsstat 1 166shows high retransmit rates while the file system is active or reducing the 167value if there is a low retransmit rate but long response delay observed. 168.It Fl w 169Set the write data size to the specified value. 170Ditto the comments w.r.t. the 171.Fl r 172option, but using the 173.Dq "fragments dropped due to timeout" 174value on the server instead of the client. 175Note that both the 176.Fl r 177and 178.Fl w 179options should only be used as a last ditch effort at improving performance 180when mounting servers that do not support TCP mounts. 181.It Fl x 182Set the retransmit timeout count for soft mounts to the specified value. 183.El 184.Sh SEE ALSO 185.Xr mount 2 , 186.Xr unmount 2 , 187.Xr fstab 5 , 188.Xr mount 8 189.Sh BUGS 190Due to the way that Sun RPC is implemented on top of UDP (unreliable datagram) 191transport, tuning such mounts is really a black art that can only be expected 192to have limited success. 193For clients mounting servers that are not on the same 194LAN cable or that tend to be overloaded, 195TCP transport is strongly recommended, 196but unfortunately this is restricted to mostly 4.4BSD servers. 197