Modify the NFSv3 and NFSv4 TCP maximum transfer size
You can modify the -tcp-max-xfer-size
option to configure maximum transfer sizes for all TCP connections using the NFSv3 and NFSv4.x protocols.
You can modify these options individually for each storage virtual machine (SVM).
Beginning with ONTAP 9, the v3-tcp-max-read-size
and v3-tcp-max-write-size
options are obsolete. You must use the -tcp-max-xfer-size
option instead.
-
Set the privilege level to advanced:
set -privilege advanced
-
Perform one of the following actions:
If you want to… Enter the command… Modify the NFSv3 or NFSv4 TCP maximum transfer size
vserver nfs modify -vserver vserver_name -tcp-max-xfer-size integer_max_xfer_size
Option Range Default -tcp-max-xfer-size
8192 to 1048576 bytes
65536 bytes
The maximum transfer size that you enter must be a multiple of 4 KB (4096 bytes). Requests that are not properly aligned negatively affect performance.
-
Use the
vserver nfs show -fields tcp-max-xfer-size
command to verify the changes. -
If any clients use static mounts, unmount and remount for the new parameter size to take effect.
The following command sets the NFSv3 and NFSv4.x TCP maximum transfer size to 1048576 bytes on the SVM named vs1:
vs1::> vserver nfs modify -vserver vs1 -tcp-max-xfer-size 1048576