Installation guide

250 Chapter 17. Network File System (NFS)
17.2.1. /etc/exports
The /etc/exports file is the standard for controlling which file systems are exported to which hosts,
as well as specifying particular options that control everything. Blank lines are ignored, comments can
be made using #, and long lines can be wrapped with a backslash (\). Each exported file system should
be on its own line. Lists of authorized hosts placed after an exported file system must be separated by
space characters. Options for each of the hosts must be placed in parentheses directly after the host
identifier, without any spaces separating the host and the first parenthesis.
In its simplest form, /etc/exports only needs to know the directory to be exported and the hosts
permitted to use it:
/some/directory bob.domain.com
/another/exported/directory 192.168.0.3
After re-exporting /etc/exports with the /sbin/service nfs reload command, the
bob.domain.com host will be able to mount /some/directory and 192.168.0.3 can mount
/another/exported/directory. Because no options are specified in this example, several
default NFS preferences take effect:
• ro — Read-only. Hosts mounting this file system will not be able to change it. To allow hosts to
make changes to the file system, you must specify rw (read-write).
• async — Allows the server to write data to the disk when it sees fit. While this is not important
if the host is accessing data as read-only, if a host is making changes to a read-write file system
and the server crashes, data could be lost. By specifying the sync option, all file writes must be
committed to the disk before the write request by the client is actually completed. This may lower
performance.
• wdelay — Causes the NFS server to delay writing to the disk if it suspects another write request
is imminent. This can improve performance by reducing the number of times the disk must be
accessed by separate write commands, reducing write overhead. Use no_wdelay to turn this feature
off, which only works if you are using the sync option.
• root_squash — Makes any client accesses to the exported file system, made as the root user on
the client machine, take place as the nobody user ID. This effectively "squashes" the power of the
remote root user to the lowest local user, preventing remote root users from acting as though they
were the root user on the local system. Alternatively, the no_root_squash option turns off root
squashing. To squash every remote user, including root, use the all_squash option. To specify
the user and group IDs to use with remote users from a particular host, use the anonuid and
anongid options, respectively. In this way, you can create a special user account for remote NFS
users to share and specify (anonuid=
uid-value ,anongid= gid-value ), where uid-
value
is the user ID number and gid-value is the group ID number.
In order to override these defaults, you must specify an option that takes its place. For example, if
you do not specify rw, then that export will only be shared read-only. Each default for every exported
file system must be explicitly overridden. Additionally, other options are available where no default
value is in place. These include the ability to disable sub-tree checking, allow access from insecure
ports, and allow insecure file locks (necessary for certain early NFS client implementations). See the
exports man page for details on these lesser used options.
When specifying hostnames, you can use the following methods:
• single host — Where one particular host is specified with a fully qualified domain name, hostname,
or IP address.
• wildcards — Where a * or ? character is used to take into account a grouping of fully qualified
domain names or IP addresses or those that match a particular string of letters.