<< Previous Page: Export NFS4 File System

nfs4-acl-tools package manages particular directories and files. It includes:

  • nfs4-getfacl: List all access controls for specified file or directory
  • nfs4-setfacl: Create access controls for a directory or a file
  • nfs4-editfacl: To edit or change the access controls. It invokes nfs_setfacl -e option and opens an editor.

An ACL entry

  • It begins with an entry type, such as accept or deny entry (A or D),
  • followed by ACL flag to specify group or inheritance capability
  • and then principal to which ACL is applied
  • and finally the list of access options like r for read and w for write

ACL Options

  • (r,w,x): read, write and execute permissions
  • (t,n) read
  • (T,N) write
  • read(c)
  • write(c)
  • C: allows user to change access controls
  • y: enables read and write synchronization
  • d: delete files and directories
  • D: for subdirectories
  • a: append data and create sub-directories
  • rtncy: all read options
  • wadDTNC: write options
  • x: execute

Special URLs OWNER@, GROUP@, and EVERYONE@ correspond to the owner, group and other access used on standard permission

Use the mount command to mount the NFSv4 file system explicitly or by creating an entry in the /etc/fstab file.

$ mount -t nfs4 wap.ananova.com:/ /home/ananova/wap
mount -t nfs4 -o soft,intr,timeo=20 wap.ananova.com:/ /home/ananova/wap

Unmount NFS directory with the unmount command

unmount /home/ananova/wap
unmount wap.ananova.com:/

An NFS entry in the /etc/fstab consists of hostname, followed by pathname of the directory, where it is mounted, separated by a colon.

host:/home/ananova home/ananova nfs options 0 0
  • rsize=n: The number of bytes NFS uses when reading files from an NFS server. The default is 1024 bytes, while the size of 8192 greatly improves performance.
  • wsize=n: The number of bytes NFS uses when writing files to an NFS server.
  • timeo=n: The value in tenths of a second before sending the first retransmission after a timeout. The default value is seven-tenths of a second.
  • retry=n: The number of minutes to retry an NFS mount operation before giving up. The default is 10,000 minutes (one week)
  • retrans=n: The number of retransmission or minor timeouts for an NFS mount operation before a major timeout. The default is 3. At that time, the connection canceled or a “server not responding” message is displayed
  • soft: Mount system using soft mount
  • hard: Mount system using hard mount. It is the default.
  • intr: Allows NFS to interrupt the file operation and return to the calling program. The default is not to allow file operations to be interrupted.
  • bg: If the first mount attempt times out continue trying the mount in the background. The default is to fall without backgrounding.
  • tcp: Mounts the NFS file system using the TCP protocol, instead of the default UDP protocol

Automount service autofs configured using the master file /etc/auto.master to list map files. The master file lists the root pathnames where the file systems mounted along with a map file for each of those pathnames.

Mac OS X for Macintosh supports NFS for file sharing.