Merge commit '850ef5ae11d69ea3381bd310f564f025fc8caea3'

[FreeBSD/FreeBSD.git] / sbin / mount_nfs / mount_nfs.8

.\" Copyright (c) 1992, 1993, 1994, 1995
.\"     The Regents of the University of California.  All rights reserved.
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\"    notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\"    notice, this list of conditions and the following disclaimer in the
.\"    documentation and/or other materials provided with the distribution.
.\" 3. Neither the name of the University nor the names of its contributors
.\"    may be used to endorse or promote products derived from this software
.\"    without specific prior written permission.
.\"
.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
.\" SUCH DAMAGE.
.\"
.Dd October 8, 2023
.Dt MOUNT_NFS 8
.Os
.Sh NAME
.Nm mount_nfs
.Nd mount NFS file systems
.Sh SYNOPSIS
.Nm
.Op Fl 23bcdiLlNPsTU
.Op Fl a Ar maxreadahead
.Op Fl D Ar deadthresh
.Op Fl g Ar maxgroups
.Op Fl I Ar readdirsize
.Op Fl o Ar options
.Op Fl R Ar retrycnt
.Op Fl r Ar readsize
.Op Fl t Ar timeout
.Op Fl w Ar writesize
.Op Fl x Ar retrans
.Ar rhost : Ns Ar path node
.Sh DESCRIPTION
The
.Nm
utility calls the
.Xr nmount 2
system call to prepare and graft a remote NFS file system
.Pq Ar rhost : Ns Ar path
on to the file system tree at the point
.Ar node .
This command is normally executed by
.Xr mount 8 .
For NFSv2 and NFSv3,
it implements the mount protocol as described in RFC 1094, Appendix A and
RFC 1813, Appendix I.
For NFSv4, it uses the NFSv4 protocol as described in RFC 7530, RFC 5661 and
RFC 7862.
.Pp
By default,
.Nm
keeps retrying until the mount succeeds.
This behaviour is intended for file systems listed in
.Xr fstab 5
that are critical to the boot process.
For non-critical file systems, the
.Cm bg
and
.Cm retrycnt
options provide mechanisms to prevent the boot process from hanging
if the server is unavailable.
.Pp
If the server becomes unresponsive while an NFS file system is
mounted, any new or outstanding file operations on that file system
will hang uninterruptibly until the server comes back.
To modify this default behaviour, see the
.Cm intr
and
.Cm soft
options.
.Pp
The options are:
.Bl -tag -width indent
.It Fl o
Options are specified with a
.Fl o
flag followed by a comma separated string of options.
See the
.Xr mount 8
man page for possible options and their meanings.
The following NFS specific options are also available:
.Bl -tag -width indent
.It Cm acregmin Ns = Ns Aq Ar seconds
.It Cm acregmax Ns = Ns Aq Ar seconds
.It Cm acdirmin Ns = Ns Aq Ar seconds
.It Cm acdirmax Ns = Ns Aq Ar seconds
When attributes of files are cached, a timeout calculated to determine
whether a given cache entry has expired.
These four values determine the upper and lower bounds of the timeouts for
.Dq directory
attributes and
.Dq regular
(ie: everything else).
The default values are 3 -> 60 seconds
for regular files, and 30 -> 60 seconds for directories.
The algorithm to calculate the timeout is based on the age of the file.
The older the file,
the longer the cache is considered valid, subject to the limits above.
.It Cm actimeo Ns = Ns Aq Ar seconds
Set four cache timeouts above to specified value.
.It Cm allgssname
This option can be used along with
.Fl o Cm gssname
to specify that all operations should use the host-based initiator
credential.
This may be used for clients that run system daemons that need to
access files on the NFSv4 mounted volume.
.It Cm bg
If an initial attempt to contact the server fails, fork off a child to keep
trying the mount in the background.
Useful for
.Xr fstab 5 ,
where the file system mount is not critical to multiuser operation.
.It Cm bgnow
Like
.Cm bg ,
fork off a child to keep trying the mount in the background,
but do not attempt to mount in the foreground first.
This eliminates a
60+ second timeout when the server is not responding.
Useful for speeding up the boot process of a client when the server is
likely to be unavailable.
This is often the case for interdependent servers
such as cross-mounted servers (each of two servers is an NFS client of
the other) and for cluster nodes that must boot before the file servers.
.It Cm deadthresh Ns = Ns Aq Ar value
Set the
.Dq "dead server threshold"
to the specified number of round trip timeout intervals before a
.Dq "server not responding"
message is displayed.
.It Cm dumbtimer
Turn off the dynamic retransmit timeout estimator.
This may be useful for UDP mounts that exhibit high retry rates,
since it is possible that the dynamically estimated timeout interval is too
short.
.It Cm fg
Same as not specifying
.Cm bg .
.It Cm gssname Ns = Ns Aq Ar service-principal-name
This option can be used with the KerberosV security flavors for NFSv4 mounts
to specify the
.Dq "service-principal-name"
of a host-based entry in the default
keytab file that is used for system operations.
It allows the mount to be performed by
.Dq "root"
and avoids problems with
cached credentials for the system operations expiring.
The
.Dq "service-principal-name"
should be specified without instance or domain and is typically
.Dq "host" ,
.Dq "nfs"
or
.Dq "root" ,
although the form
.Sm off
.Aq Ar service
@
.Aq Ar fqdn
.Sm on
can also be used if the local system's
.Xr gethostname 3
value does not match the host-based principal in the keytab.
.It Cm hard
Same as not specifying
.Cm soft .
.It Cm intr
Make the mount interruptible, which implies that file system calls that
are delayed due to an unresponsive server will fail with EINTR when a
termination signal is posted for the process.
To avoid leaving file locks in an indeterminate state on the NFS
server, it is recommended that the
.Cm nolockd
option be used with this option.
.It Cm maxgroups Ns = Ns Aq Ar value
Set the maximum size of the group list for the credentials to the
specified value.
This should be used for mounts on old servers that cannot handle a
group list size of 16, as specified in RFC 1057.
Try 8, if users in a lot of groups cannot get response from the mount
point.
.It Cm mntudp
Force the mount protocol to use UDP transport, even for TCP NFS mounts.
(Necessary for some old
.Bx
servers.)
.It Cm nametimeo Ns = Ns Aq Ar value
Override the default of NFS_DEFAULT_NAMETIMEO for the timeout (in seconds)
for positive name cache entries.
If this is set to 0 it disables positive name caching for the mount point.
.It Cm negnametimeo Ns = Ns Aq Ar value
Override the default of NFS_DEFAULT_NEGNAMETIMEO for the timeout (in seconds)
for negative name cache entries.
If this is set to 0 it disables negative name caching for the mount point.
.It Cm nconnect Ns = Ns Aq Ar value
Specify the number of TCP connections (1-16) to be used
for an NFS Version 4, minor version 1 or 2 mount.
Multiple TCP connections can provide more client to server network
bandwidth for certain network configurations such as:
.Bd -literal
- Multiple network interfaces that are aggregated together.
- A fast network interface that uses multiple queues.
.Ed
.sp
The first TCP connection will be used for all RPCs that consist
entirely of small RPC messages.
The RPCs that can have large RPC messages (Read/Readdir/Write) are
distributed over the additional TCP connections in a round robin
fashion.
This option will result in more IP port#s being used.
This option requires the
.Cm nfsv4
option.
Note that for NFS servers such as AmazonEFS, where each new TCP
connection can connect to a different cluster that maintains lock
state separately, this option cannot be used.
.It Cm nfsv2
Use the NFS Version 2 protocol (the default is to try version 3 first
then version 2).
Note that NFS version 2 has a file size limit of 2 gigabytes.
.It Cm nfsv3
Use the NFS Version 3 protocol.
.It Cm nfsv4
Use the NFS Version 4 protocol.
This option will force the mount to use
TCP transport.
By default, the highest minor version of NFS Version 4 that is
supported by the NFS Version 4 server will be used.
See the
.Cm minorversion
option.
Make sure that all your NFS Version 4 clients have unique
values in
.Pa /etc/hostid .
.It Cm minorversion Ns = Ns Aq Ar value
Use the specified minor version for a NFS Version 4 mount,
overriding the default.
The minor versions supported are 0, 1, and 2.
This option is only meaningful when used with the
.Cm nfsv4
option.
.It Cm oneopenown
Make a minor version 1 or 2 of the NFS Version 4 protocol mount use a single
OpenOwner for all Opens.
This may be useful for a server with a very low limit on OpenOwners, such as
AmazonEFS.
It may be required when an accumulation of NFS version 4 Opens occurs,
as indicated by the
.Dq Opens
count displayed by
.Xr nfsstat 1
with the
.Fl c
and
.Fl E
command-line options.
A common case for an accumulation of Opens is a shared library within
the NFS mount that is used by several
processes, where at least one of these processes is always running.
This option cannot be used for an NFS Version 4, minor version 0 mount.
It may not work correctly when Delegations are being issued by a server,
but note that the AmazonEFS server does not issued delegations at this time.
This option is only meaningful when used with the
.Cm nfsv4
option.
.It Cm pnfs
Enable support for parallel NFS (pNFS) for minor version 1 or 2 of the
NFS Version 4 protocol.
This option is only meaningful when used with the
.Cm nfsv4
option.
.It Cm noac
Disable attribute caching.
.It Cm noconn
For UDP mount points, do not do a
.Xr connect 2 .
This must be used if the server does not reply to requests from the standard
NFS port number 2049 or replies to requests using a different IP address
(which can occur if the server is multi-homed).
Setting the
.Va vfs.nfs.nfs_ip_paranoia
sysctl to 0 will make this option the default.
.It Cm nocto
Normally, NFS clients maintain the close-to-open cache coherency.
This works by flushing at close time and checking at open time.
Checking at open time is implemented by getting attributes from
the server and purging the data cache if they do not match
attributes cached by the client.
.Pp
This option disables checking at open time.
It may improve performance for read-only mounts,
but should only be used if the data on the server changes rarely.
Be sure to understand the consequences before enabling this option.
.It Cm noinet4 , noinet6
Disables
.Dv AF_INET
or
.Dv AF_INET6
connections.
Useful for hosts that have
both an A record and an AAAA record for the same name.
.It Cm nolockd
Do
.Em not
forward
.Xr fcntl 2
locks over the wire via the NLM protocol for NFSv3 mounts
or via the NFSv4 protocol for NFSv4 mounts.
All locks will be local and not seen by the server
and likewise not seen by other NFS clients for NFSv3 or NFSv4 mounts.
This removes the need to run the
.Xr rpcbind 8
service and the
.Xr rpc.statd 8
and
.Xr rpc.lockd 8
servers on the client for NFSv3 mounts.
Note that this option will only be honored when performing the
initial mount, it will be silently ignored if used while updating
the mount options.
Also, note that NFSv4 mounts do not use these daemons.
The NFSv4 protocol handles locks,
unless this option is specified.
.It Cm noncontigwr
This mount option allows the NFS client to
combine non-contiguous byte ranges being written
such that the dirty byte range becomes a superset of the bytes
that are dirty.
This reduces the number of writes significantly for software
builds.
The merging of byte ranges is not done if the file has been file
locked, since most applications modifying a file from multiple
clients will use file locking.
As such, this option could result in a corrupted file for the
rare case of an application modifying the file from multiple
clients concurrently without using file locking.
.It Cm principal
For the RPCSEC_GSS security flavors, such as krb5, krb5i and krb5p,
this option sets the name of the host based principal name expected
by the server.
This option overrides the default, which will be ``nfs@<server-fqdn>''
and should normally be sufficient.
.It Cm noresvport
Do
.Em not
use a reserved socket port number (see below).
.It Cm port Ns = Ns Aq Ar port_number
Use specified port number for NFS requests.
The default is to query the portmapper for the NFS port.
.It Cm proto Ns = Ns Aq Ar protocol
Specify transport protocol version to use.
Currently, they are:
.Bd -literal
udp -   Use UDP over IPv4
tcp -   Use TCP over IPv4
udp6 -  Use UDP over IPv6
tcp6 -  Use TCP over IPv6
.Ed
.It Cm rdirplus
Used with NFSV3 to specify that the \fBReaddirPlus\fR RPC should
be used.
For NFSV4, setting this option has a similar effect, in that it will make
the Readdir Operation get more attributes.
This option reduces RPC traffic for cases such as
.Dq "ls -l" ,
but tends to flood the attribute and name caches with prefetched entries.
Try this option and see whether performance improves or degrades.
Probably
most useful for client to server network interconnects with a large bandwidth
times delay product.
.It Cm readahead Ns = Ns Aq Ar value
Set the read-ahead count to the specified value.
This may be in the range of 0 - 4, and determines how many blocks
will be read ahead when a large file is being read sequentially.
Trying a value greater than 1 for this is suggested for
mounts with a large bandwidth * delay product.
.It Cm readdirsize Ns = Ns Aq Ar value
Set the readdir read size to the specified value.
The value should normally
be a multiple of
.Dv DIRBLKSIZ
that is <= the read size for the mount.
.It Cm resvport
Use a reserved socket port number.
This flag is obsolete, and only retained for compatibility reasons.
Reserved port numbers are used by default now.
(For the rare case where the client has a trusted root account
but untrustworthy users and the network cables are in secure areas this does
help, but for normal desktop clients this does not apply.)
.It Cm retrans Ns = Ns Aq Ar value
Set the retransmit timeout count for soft mounts to the specified value.
.It Cm retrycnt Ns = Ns Aq Ar count
Set the mount retry count to the specified value.
The default is a retry count of zero, which means to keep retrying
forever.
There is a 60 second delay between each attempt.
.It Cm rsize Ns = Ns Aq Ar value
Set the read data size to the specified value.
It should normally be a power of 2 greater than or equal to 1024.
This should be used for UDP mounts when the
.Dq "fragments dropped due to timeout"
value is getting large while actively using a mount point.
(Use
.Xr netstat 1
with the
.Fl s
option to see what the
.Dq "fragments dropped due to timeout"
value is.)
.It Cm sec Ns = Ns Aq Ar flavor
This option specifies what security flavor should be used for the mount.
Currently, they are:
.Bd -literal
krb5 -  Use KerberosV authentication
krb5i - Use KerberosV authentication and
        apply integrity checksums to RPCs
krb5p - Use KerberosV authentication and
        encrypt the RPC data
sys -   The default AUTH_SYS, which uses a
        uid + gid list authenticator
.Ed
.It Cm soft
A soft mount, which implies that file system calls will fail
after
.Ar retrycnt
round trip timeout intervals.
.It Cm syskrb5
This option specifies that a KerberosV NFSv4 minor version 1 or 2 mount
uses AUTH_SYS for system operations.
Using this option avoids the need for a KerberosV mount to have a
host-based principal entry in the default keytab file
(no
.Cm gssname
option) or a requirement for the user doing the mount to have a
valid KerberosV ticket granting ticket (TGT) when the mount is done.
This option is intended to be used with the
.Cm sec Ns = Ns krb5
and
.Cm tls
options and can only be used for
NFSv4 mounts with minor version 1 or 2.
.It Cm tcp
Use TCP transport.
This is the default option, as it provides for increased reliability on both
LAN and WAN configurations compared to UDP.
Some old NFS servers do not support this method; UDP mounts may be required
for interoperability.
.It Cm timeout Ns = Ns Aq Ar value
Set the initial retransmit timeout to the specified value,
expressed in tenths of a second.
May be useful for fine tuning UDP mounts over internetworks
with high packet loss rates or an overloaded server.
Try increasing the interval if
.Xr nfsstat 1
shows high retransmit rates while the file system is active or reducing the
value if there is a low retransmit rate but long response delay observed.
(Normally, the
.Cm dumbtimer
option should be specified when using this option to manually
tune the timeout
interval.)
.It Cm timeo Ns = Ns Aq Ar value
Alias for
.Cm timeout .
.It Cm tls
This option specifies that the connection to the server must use TLS
per RFC 9289.
TLS is only supported for TCP connections and the
.Xr rpc.tlsclntd 8
daemon must be running for an NFS over TCP connection to use TLS.
.It Cm tlscertname Ns = Ns Aq Ar name
This option specifies the name of an alternate certificate to be
presented to the NFS server during TLS handshake.
The default certificate file names are
.Dq cert.pem
and
.Dq certkey.pem .
When this option is specified,
.Ar name
replaces
.Dq cert
in the above file names.
For example, if the value of
.Ar name
is specified as
.Dq other
the certificate file names to be used will be
.Dq other.pem
and
.Dq otherkey.pem .
These files are stored in
.Pa /etc/rpc.tlsclntd
by default.
This option is only meaningful when used with the
.Cm tls
option and the
.Xr rpc.tlsclntd 8
is running with the
.Fl m
command line flag set.
.It Cm udp
Use UDP transport.
.It Cm vers Ns = Ns Aq Ar vers_number
Use the specified version number for NFS requests.
See the
.Cm nfsv2 ,
.Cm nfsv3 ,
and
.Cm nfsv4
options for details.
.It Cm wcommitsize Ns = Ns Aq Ar value
Set the maximum pending write commit size to the specified value.
This determines the maximum amount of pending write data that the NFS
client is willing to cache for each file.
.It Cm wsize Ns = Ns Aq Ar value
Set the write data size to the specified value.
Ditto the comments w.r.t.\& the
.Cm rsize
option, but using the
.Dq "fragments dropped due to timeout"
value on the server instead of the client.
Note that both the
.Cm rsize
and
.Cm wsize
options should only be used as a last ditch effort at improving performance
when mounting servers that do not support TCP mounts.
.El
.El
.Sh IMPLEMENTATION NOTES
When neither the
.Cm rsize
nor
.Cm wsize
options are specified, the I/O size will be set to the largest value
supported by both the NFS client and server.
The largest value supported by the NFS client is defined by
the tunable
.Cd vfs.maxbcachebuf
which can be set to a power of two up to
.Cd kern.maxphys .
.Pp
The
.Xr nfsstat 1
command with the
.Ic -m
command line option will show what
.Nm
option settings are actually in use for the mount.
.Sh COMPATIBILITY
The following command line flags are equivalent to
.Fl o
named options and are supported for compatibility with older
installations.
.Bl -tag -width indent
.It Fl 2
Same as
.Fl o Cm nfsv2
.It Fl 3
Same as
.Fl o Cm nfsv3
.It Fl D
Same as
.Fl o Cm deadthresh
.It Fl I
Same as
.Fl o Cm readdirsize Ns = Ns Aq Ar value
.It Fl L
Same as
.Fl o Cm nolockd
.It Fl N
Same as
.Fl o Cm noresvport
.It Fl P
Use a reserved socket port number.
This flag is obsolete, and only retained for compatibility reasons.
(For the rare case where the client has a trusted root account
but untrustworthy users and the network cables are in secure areas this does
help, but for normal desktop clients this does not apply.)
.It Fl R
Same as
.Fl o Cm retrycnt Ns = Ns Aq Ar value
.It Fl T
Same as
.Fl o Cm tcp
.It Fl U
Same as
.Fl o Cm mntudp
.It Fl a
Same as
.Fl o Cm readahead Ns = Ns Aq Ar value
.It Fl b
Same as
.Fl o Cm bg
.It Fl c
Same as
.Fl o Cm noconn
.It Fl d
Same as
.Fl o Cm dumbtimer
.It Fl g
Same as
.Fl o Cm maxgroups
.It Fl i
Same as
.Fl o Cm intr
.It Fl l
Same as
.Fl o Cm rdirplus
.It Fl r
Same as
.Fl o Cm rsize Ns = Ns Aq Ar value
.It Fl s
Same as
.Fl o Cm soft
.It Fl t
Same as
.Fl o Cm retransmit Ns = Ns Aq Ar value
(deprecated)
.It Fl w
Same as
.Fl o Cm wsize Ns = Ns Aq Ar value
.It Fl x
Same as
.Fl o Cm retrans Ns = Ns Aq Ar value
.El
.Pp
The following
.Fl o
named options are equivalent to other
.Fl o
named options and are supported for compatibility with other
operating systems (e.g., Linux, Solaris, and OSX) to ease usage of
.Xr autofs 5
support.
.Bl -tag -width indent
.It Fl o Cm vers Ns = Ns 2
Same as
.Fl o Cm nfsv2
.It Fl o Cm vers Ns = Ns 3
Same as
.Fl o Cm nfsv3
.It Fl o Cm vers Ns = Ns 4
Same as
.Fl o Cm nfsv4
.El
.Sh SEE ALSO
.Xr nfsstat 1 ,
.Xr nmount 2 ,
.Xr unmount 2 ,
.Xr lagg 4 ,
.Xr nfsv4 4 ,
.Xr fstab 5 ,
.Xr gssd 8 ,
.Xr mount 8 ,
.Xr nfsd 8 ,
.Xr nfsiod 8 ,
.Xr rpc.tlsclntd 8 ,
.Xr showmount 8
.Sh HISTORY
A version of the
.Nm
utility appeared in
.Bx 4.4 .
.Sh BUGS
Since NFSv4 performs open/lock operations that have their ordering strictly
enforced by the server, the options
.Cm intr
and
.Cm soft
cannot be safely used.
For NFSv4 minor version 1 or 2 mounts, the ordering is done
via session slots and the NFSv4 client now handles broken session slots
fairly well.
As such, if the
.Cm nolockd
option is used along with
.Cm intr
and/or
.Cm soft ,
an NFSv4 minor version 1 or 2 mount
should work fairly well, although still not completely correctly.
For NFSv4 minor version 0 mounts,
.Cm hard
mounts without the
.Cm intr
mount option is strongly recommended.