Mount NFS File System Over a Slow and Busy Network

Posted on in Categories , , , , , , , last updated September 7, 2011

Traffic between my home and office data center encrypted using VPN but sometime my ISP get little slow or network becomes busy. I’m using Fedora Core Linux desktop system and my system freezes for a few minutes. What optimization options should be used when mounting an exported NFS share over a slow and busy network?

Generally, you will see server not responding message or computer just freezes. You can pass the following option while mounting remote shared/exported NFS server:

  1. soft: This option allows the kernel to time out if the nfs server is not responding for some time. The time can be specified with timeo=time. This option might be useful if your nfs server sometimes doesn’t respond or will be rebooted while some process tries to get a file from the server.
  2. timeo N: This timeout value is expressed in tenths of a second
  3. retrans : The number of timeouts allowed before the client gives up and displays the Server not responding message. The mount fails soon after.
  4. proto tcp/udp: The protocol (Either TCP or UDP) used when attempting to mount the server.

For example:
# mount -t nfs -o soft,timeo=900,retrans=3,vers=3, proto=tcp remotenfsserver:/home/data /mnt/nfs-mnt-point
# mount.nfs4 -o soft,timeo=900,retrans=3,vers=3, proto=tcp nas01:/data /data

2 comment

  1. The Linux manpage for MOUNT_NFS(8) specifically states that changing the timeo value is useful when optimizing UDP mounts. If the timeo value only affects UDP mounts specifying “proto=tcp” would make the value for timeo moot would it not?

Leave a Comment