1.\" Copyright (c) 1989, 1991, 1993 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.\" 3. 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.\" @(#)nfsd.8 8.4 (Berkeley) 3/29/95 29.\" $FreeBSD$ 30.\" 31.Dd August 5, 2018 32.Dt NFSD 8 33.Os 34.Sh NAME 35.Nm nfsd 36.Nd remote 37.Tn NFS 38server 39.Sh SYNOPSIS 40.Nm 41.Op Fl ardute 42.Op Fl n Ar num_servers 43.Op Fl h Ar bindip 44.Op Fl p Ar pnfs_setup 45.Op Fl m Ar mirror_level 46.Op Fl Fl maxthreads Ar max_threads 47.Op Fl Fl minthreads Ar min_threads 48.Sh DESCRIPTION 49The 50.Nm 51utility runs on a server machine to service 52.Tn NFS 53requests from client machines. 54At least one 55.Nm 56must be running for a machine to operate as a server. 57.Pp 58Unless otherwise specified, eight servers per CPU for 59.Tn UDP 60transport are started. 61.Pp 62The following options are available: 63.Bl -tag -width Ds 64.It Fl r 65Register the 66.Tn NFS 67service with 68.Xr rpcbind 8 69without creating any servers. 70This option can be used along with the 71.Fl u 72or 73.Fl t 74options to re-register NFS if the rpcbind server is restarted. 75.It Fl d 76Unregister the 77.Tn NFS 78service with 79.Xr rpcbind 8 80without creating any servers. 81.It Fl n Ar threads 82Specifies how many servers to create. This option is equivalent to specifying 83.Fl Fl maxthreads 84and 85.Fl Fl minthreads 86with their respective arguments to 87.Ar threads . 88.It Fl Fl maxthreads Ar threads 89Specifies the maximum servers that will be kept around to service requests. 90.It Fl Fl minthreads Ar threads 91Specifies the minimum servers that will be kept around to service requests. 92.It Fl h Ar bindip 93Specifies which IP address or hostname to bind to on the local host. 94This option is recommended when a host has multiple interfaces. 95Multiple 96.Fl h 97options may be specified. 98.It Fl a 99Specifies that nfsd should bind to the wildcard IP address. 100This is the default if no 101.Fl h 102options are given. 103It may also be specified in addition to any 104.Fl h 105options given. 106Note that NFS/UDP does not operate properly when 107bound to the wildcard IP address whether you use -a or do not use -h. 108.It Fl p Ar pnfs_setup 109Enables pNFS support in the server and specifies the information that the 110daemon needs to start it. 111This option can only be used on one server and specifies that this server 112will be the MetaData Server (MDS) for the pNFS service. 113This can only be done if there is at least one FreeBSD system configured 114as a Data Server (DS) for it to use. 115.Pp 116The 117.Ar pnfs_setup 118string is a set of fields separated by ',' characters: 119.Bl -tag -width Ds 120Each of these fields specifies one DS. 121It consists of a server hostname, followed by a ':' 122and the directory path where the DS's data storage file system is mounted on 123this MDS server. 124This can optionally be followed by a '#' and the mds_path, which is the 125directory path for an exported file system on this MDS. 126If this is specified, it means that this DS is to be used to store data 127files for this mds_path file system only. 128If this optional component does not exist, the DS will be used to store data 129files for all exported MDS file systems. 130The DS storage file systems must be mounted on this system before the 131.Nm 132is started with this option specified. 133.br 134For example: 135.sp 136nfsv4-data0:/data0,nfsv4-data1:/data1 137.sp 138would specify two DS servers called nfsv4-data0 and nfsv4-data1 that comprise 139the data storage component of the pNFS service. 140These two DSs would be used to store data files for all exported file systems 141on this MDS. 142The directories 143.Dq /data0 144and 145.Dq /data1 146are where the data storage servers exported 147storage directories are mounted on this system (which will act as the MDS). 148.br 149Whereas, for the example: 150.sp 151nfsv4-data0:/data0#/export1,nfsv4-data1:/data1#/export2 152.sp 153would specify two DSs as above, however nfsv4-data0 will be used to store 154data files for 155.Dq /export1 156and nfsv4-data1 will be used to store data files for 157.Dq /export2 . 158.El 159.sp 160When using IPv6 addresses for DSs 161be wary of using link local addresses. 162The IPv6 address for the DS is sent to the client and there is no scope 163zone in it. 164As such, a link local address may not work for a pNFS client to DS 165TCP connection. 166When parsed, 167.Nm 168will only use a link local address if it is the only address returned by 169.Xr getaddrinfo 3 170for the DS hostname. 171.It Fl m Ar mirror_level 172This option is only meaningful when used with the 173.Fl p 174option. 175It specifies the 176.Dq mirror_level , 177which defines how many of the DSs will 178have a copy of a file's data storage file. 179The default of one implies no mirroring of data storage files on the DSs. 180The 181.Dq mirror_level 182would normally be set to 2 to enable mirroring, but 183can be as high as NFSDEV_MAXMIRRORS. 184There must be at least 185.Dq mirror_level 186DSs for each exported file system on the MDS, as specified in the 187.Fl p 188option. 189This implies that, for the above example using "#/export1" and "#/export2", 190mirroring cannot be done. 191There would need to be two DS entries for each of "#/export1" and "#/export2" 192in order to support a 193.Dq mirror_level 194of two. 195.Pp 196If mirroring is enabled, the server must use the Flexible File 197layout. 198If mirroring is not enabled, the server will use the File layout 199by default, but this default can be changed to the Flexible File layout if the 200.Xr sysctl 1 201vfs.nfsd.default_flexfile 202is set non-zero. 203.It Fl t 204Serve 205.Tn TCP NFS 206clients. 207.It Fl u 208Serve 209.Tn UDP NFS 210clients. 211.It Fl e 212Ignored; included for backward compatibility. 213.El 214.Pp 215For example, 216.Dq Li "nfsd -u -t -n 6" 217serves 218.Tn UDP 219and 220.Tn TCP 221transports using six daemons. 222.Pp 223A server should run enough daemons to handle 224the maximum level of concurrency from its clients, 225typically four to six. 226.Pp 227The 228.Nm 229utility listens for service requests at the port indicated in the 230.Tn NFS 231server specification; see 232.%T "Network File System Protocol Specification" , 233RFC1094, 234.%T "NFS: Network File System Version 3 Protocol Specification" , 235RFC1813, 236.%T "Network File System (NFS) Version 4 Protocol" , 237RFC3530 and 238.%T "Network File System (NFS) Version 4 Minor Version 1 Protocol" , 239RFC5661. 240.Pp 241If 242.Nm 243detects that 244.Tn NFS 245is not loaded in the running kernel, it will attempt 246to load a loadable kernel module containing 247.Tn NFS 248support using 249.Xr kldload 2 . 250If this fails, or no 251.Tn NFS 252KLD is available, 253.Nm 254will exit with an error. 255.Pp 256If 257.Nm 258is to be run on a host with multiple interfaces or interface aliases, use 259of the 260.Fl h 261option is recommended. 262If you do not use the option NFS may not respond to 263UDP packets from the same IP address they were sent to. 264Use of this option 265is also recommended when securing NFS exports on a firewalling machine such 266that the NFS sockets can only be accessed by the inside interface. 267The 268.Nm ipfw 269utility 270would then be used to block nfs-related packets that come in on the outside 271interface. 272.Pp 273If the server has stopped servicing clients and has generated a console message 274like 275.Dq Li "nfsd server cache flooded..." , 276the value for vfs.nfsd.tcphighwater needs to be increased. 277This should allow the server to again handle requests without a reboot. 278Also, you may want to consider decreasing the value for 279vfs.nfsd.tcpcachetimeo to several minutes (in seconds) instead of 12 hours 280when this occurs. 281.Pp 282Unfortunately making vfs.nfsd.tcphighwater too large can result in the mbuf 283limit being reached, as indicated by a console message 284like 285.Dq Li "kern.ipc.nmbufs limit reached" . 286If you cannot find values of the above 287.Nm sysctl 288values that work, you can disable the DRC cache for TCP by setting 289vfs.nfsd.cachetcp to 0. 290.Pp 291The 292.Nm 293utility has to be terminated with 294.Dv SIGUSR1 295and cannot be killed with 296.Dv SIGTERM 297or 298.Dv SIGQUIT . 299The 300.Nm 301utility needs to ignore these signals in order to stay alive as long 302as possible during a shutdown, otherwise loopback mounts will 303not be able to unmount. 304If you have to kill 305.Nm 306just do a 307.Dq Li "kill -USR1 <PID of master nfsd>" 308.Sh EXIT STATUS 309.Ex -std 310.Sh SEE ALSO 311.Xr nfsstat 1 , 312.Xr kldload 2 , 313.Xr nfssvc 2 , 314.Xr nfsv4 4 , 315.Xr pnfs 4 , 316.Xr pnfsserver 4 , 317.Xr exports 5 , 318.Xr stablerestart 5 , 319.Xr gssd 8 , 320.Xr ipfw 8 , 321.Xr mountd 8 , 322.Xr nfsiod 8 , 323.Xr nfsrevoke 8 , 324.Xr nfsuserd 8 , 325.Xr rpcbind 8 326.Sh HISTORY 327The 328.Nm 329utility first appeared in 330.Bx 4.4 . 331.Sh BUGS 332If 333.Nm 334is started when 335.Xr gssd 8 336is not running, it will service AUTH_SYS requests only. To fix the problem 337you must kill 338.Nm 339and then restart it, after the 340.Xr gssd 8 341is running. 342.Pp 343If mirroring is enabled via the 344.Fl m 345option and there are Linux clients doing NFSv4.1 mounts, those clients 346need to be patched to support the 347.Dq tightly coupled 348variant of 349the Flexible File layout or the 350.Xr sysctl 1 351vfs.nfsd.flexlinuxhack 352must be set to one on the MDS as a workaround. 353