1. Network File System CIS 238

2. NFS (Network File System) The most commercially successful and widely available remote file system protocol Designed and implemented by Sun Microsystems (Walash et al, 1985; Sandberg et al, 1985) Reasons for success The NFS protocol is public domain Sun sells that implementation to all people for less than the cost of implementing it themselves Evolved from version 2 to version 3 (which is the common implementation today).

3. NFS Overview Views a set of interconnected workstations as a set of independent machines with independent file systems The goal is to allow some degree of sharing among these file systems (on explicit request) Sharing is based on client server relationships A machine may be both client and server The protocol is stateless Designed to support UNIX file system semantics The protocol design is transport independent

4. NFS Services The NFS server package includes three facilities, included in the portmap and nfs-utils packages. portmap : It maps calls made from other machines to the correct RPC service (not required with NFSv4). nfs: It translates remote file sharing requests into requests on the local file system. rpc.mountd: This service is responsible for mounting

5. NFS Configuration Files /etc/exports : Its a main configuration file of NFS, all exported files and directories are defined in this file at the NFS Server end. /etc/fstab : To mount a NFS directory on your system across the reboots, we need to make an entry in /etc/fstab. /etc/sysconfig/nfs - or /etc/nfsconfig: Configuration file of NFS to control on which port rpc and other services are listening and the NFS version used.

6. Server Installation + Setup yum install nfs-utils nfs-utils-lib yum install portmap (NFSv2 or NFSv3) -OR- yum install rpcbind( NFSv4) /etc/init.d/portmap start -OR- /etc/init.d/rpcbind start /etc/init.d/nfs start chkconfig --level 35 portmap (rpcbond) on [chkconfig --level 35 nfs on /etc/exports /nfsshare 172.27.0.0(rw,sync,no_root_squash) exportfs -a : Exports all shares listed in /etc/exports, or given name exportfs -v : Displays a list of shares files and options on a server exportfs -u : Unexports all shares listed in /etc/exports, or given name exportfs -r : Refresh the server’s list after modifying /etc/exports

7. NFS Options ro: With the help of this option we can provide read only access to the shared files i.e client will only be able to read. rw: This option allows the client server to both read and write access within the shared directory. sync: Sync confirms requests to the shared directory only once the changes have been committed. no_subtree_check: This option prevents the subtree checking. When a shared directory is the subdirectory of a larger file system, nfs performs scans of every directory above it, in order to verify its permissions and details. Disabling the subtree check may increase the reliability of NFS, but reduce security. no_root_squash: This phrase allows root to connect to the designated directory.

8. NFS Client showmount -e showmount –d mount -t nfs :/nfsshare /mnt/nfsshare /etc/fstab - :/nfsshare /mnt nfs defaults 0 0 See text or man page for mount options df -h

9. RPC(BIND) Transport: both TCP and UDP Data types: atomic types and non-recursive structures Pointers are not supported Complex memory objects (e.g., linked lists) are not supported NFS is built on top of RPC

10. The mount protocol Is used to establish the initial logical connection between a server and a client Each machine has a server process (daemon) (outside the kernel) performing the protocol functions The server has a list of exported directories (/etc/exports) The portmap service is used to find the location (port number) of the server mount service

11. The mount protocol Client’s mount process send message to the server’s portmap (or RPCBIND) daemon requesting port number of the server’s mountd daemon Server’s portmap (or RPCBIND) daemon returns the requested info Client’s mountd send the server’s mountd a request with the path of the flie system it wants to mount Server’s mountd request a file handle from the kernel - If the request is successful the handle is returned to the client - If not error is returned The client’s mountd perform the mount() system call using the received file handle