Chapter 10. Network File System (NFS)

A Network File System (NFS) allows remote hosts to mount file systems over a network and interact with those file systems as though they are mounted locally. This enables system administrators to consolidate resources onto centralized servers on the network.

This chapter focuses on fundamental NFS concepts and supplemental information.

10.1. How It Works

Currently, there are three versions of NFS. NFS version 2 (NFSv2) is older and is widely supported. NFS version 3 (NFSv3) supports safe asynchronous writes and a more robust error handling than NFSv2; it also supports 64-bit file sizes and offsets, allowing clients to access more than 2Gb of file data.

NFS version 4 (NFSv4) works through firewalls and on the Internet, no longer requires an rpcbind service, supports ACLs, and utilizes stateful operations. Red Hat Enterprise Linux supports NFSv2, NFSv3, and NFSv4 clients. When mounting a file system via NFS, Red Hat Enterprise Linux uses NFSv4 by default, if the server supports it.

All versions of NFS can use Transmission Control Protocol (TCP) running over an IP network, with NFSv4 requiring it. NFSv2 and NFSv3 can use the User Datagram Protocol (UDP) running over an IP network to provide a stateless network connection between the client and server.

When using NFSv2 or NFSv3 with UDP, the stateless UDP connection (under normal conditions) has less protocol overhead than TCP. This can translate into better performance on very clean, non-congested networks. However, because UDP is stateless, if the server goes down unexpectedly, UDP clients continue to saturate the network with requests for the server. In addition, when a frame is lost with UDP, the entire RPC request must be retransmitted; with TCP, only the lost frame needs to be resent. For these reasons, TCP is the preferred protocol when connecting to an NFS server.

The mounting and locking protocols have been incorporated into the NFSv4 protocol. The server also listens on the well-known TCP port 2049. As such, NFSv4 does not need to interact with rpcbind^[2], rpc.lockd, and rpc.statd daemons. The rpc.mountd daemon is still required on the NFS server so set up the exports, but is not involved in any over-the-wire operations.

Note

TCP is the default transport protocol for NFS version 2 and 3 under Red Hat Enterprise Linux. UDP can be used for compatibility purposes as needed, but is not recommended for wide usage. NFSv4 requires TCP.

All the RPC/NFS daemon have a '-p' command line option that can set the port, making firewall configuration easier.

After TCP wrappers grant access to the client, the NFS server refers to the /etc/exports configuration file to determine whether the client is allowed to access any exported file systems. Once verified, all file and directory operations are available to the user.

Important

In order for NFS to work with a default installation of Red Hat Enterprise Linux with a firewall enabled, configure IPTables with the default TCP port 2049. Without proper IPTables configuration, NFS will not function properly.

The NFS initialization script and rpc.nfsd process now allow binding to any specified port during system start up. However, this can be error-prone if the port is unavailable, or if it conflicts with another daemon.

10.1.1. Required Services

Red Hat Enterprise Linux uses a combination of kernel-level support and daemon processes to provide NFS file sharing. All NFS versions rely on Remote Procedure Calls (RPC) between clients and servers. RPC services under Red Hat Enterprise Linux 6 are controlled by the rpcbind service. To share or mount NFS file systems, the following services work together, depending on which version of NFS is implemented:

Note

The portmap service was used to map RPC program numbers to IP address port number combinations in earlier versions of Red Hat Enterprise Linux. This service is now replaced by rpcbind in Red Hat Enterprise Linux 6 to enable IPv6 support. For more information about this change, refer to the following links:

TI-RPC / rpcbind support: http://nfsv4.bullopensource.org/doc/tirpc_rpcbind.php
IPv6 support in NFS: http://nfsv4.bullopensource.org/doc/nfs_ipv6.php

nfs: service nfs start starts the NFS server and the appropriate RPC processes to service requests for shared NFS file systems.
nfslock: service nfslock start activates a mandatory service that starts the appropriate RPC processes which allow NFS clients to lock files on the server.
rpcbind: rpcbind accepts port reservations from local RPC services. These ports are then made available (or advertised) so the corresponding remote RPC services can access them. rpcbind responds to requests for RPC services and sets up connections to the requested RPC service. This is not used with NFSv4.

The following RPC processes facilitate NFS services:

rpc.mountd: This process receives mount requests from NFS clients and verifies that the requested file system is currently exported. This process is started automatically by the nfs service and does not require user configuration.
rpc.nfsd: rpc.nfsd allows explicit NFS versions and protocols the server advertises to be defined. It works with the Linux kernel to meet the dynamic demands of NFS clients, such as providing server threads each time an NFS client connects. This process corresponds to the nfs service.
rpc.lockd: rpc.lockd allows NFS clients to lock files on the server. If rpc.lockd is not started, file locking will fail. rpc.lockd implements the Network Lock Manager (NLM) protocol. This process corresponds to the nfslock service. This is not used with NFSv4.
rpc.statd: This process implements the Network Status Monitor (NSM) RPC protocol, which notifies NFS clients when an NFS server is restarted without being gracefully brought down. rpc.statd is started automatically by the nfslock service, and does not require user configuration. This is not used with NFSv4.
rpc.rquotad: This process provides user quota information for remote users. rpc.rquotad is started automatically by the nfs service and does not require user configuration.
rpc.idmapd: rpc.idmapd provides NFSv4 client and server upcalls, which map between on-the-wire NFSv4 names (which are strings in the form of user@domain) and local UIDs and GIDs. For idmapd to function with NFSv4, the /etc/idmapd.conf must be configured. This service is required for use with NFSv4, although not when all hosts share the same DNS domain name.

^[2] The rpcbind service replaces portmap, which was used in previous versions of Red Hat Enterprise Linux to map RPC program numbers to IP address port number combinations. For more information, refer to Section 10.1.1, “Required Services”.