Document the use of the vfs.nfsd sysctls that control the size of

the NFS server's DRC for TCP.
This is a content change.

MFC after:	1 week
This commit is contained in:
Rick Macklem 2014-08-10 20:05:13 +00:00
parent 19e2ce2d83
commit fd71d3ebee
Notes: svn2git 2020-12-20 02:59:44 +00:00
svn path=/head/; revision=269788

View File

@ -28,7 +28,7 @@
.\" @(#)nfsd.8 8.4 (Berkeley) 3/29/95
.\" $FreeBSD$
.\"
.Dd July 18, 2014
.Dd August 10, 2014
.Dt NFSD 8
.Os
.Sh NAME
@ -175,6 +175,24 @@ utility
would then be used to block nfs-related packets that come in on the outside
interface.
.Pp
If the server has stopped servicing clients and has generated a console message
like
.Dq Li "nfsd server cache flooded..." ,
the value for vfs.nfsd.tcphighwater needs to be increased.
This should allow the server to again handle requests without a reboot.
Also, you may want to consider decreasing the value for
vfs.nfsd.tcpcachetimeo to several minutes (in seconds) instead of 12 hours
when this occurs.
.Pp
Unfortunately making vfs.nfsd.tcphighwater too large can result in the mbuf
limit being reached, as indicated by a console message
like
.Dq Li "kern.ipc.nmbufs limit reached" .
If you cannot find values of the above
.Nm sysctl
values that work, you can disable the DRC cache for TCP by setting
vfs.nfsd.cachetcp to 0.
.Pp
The
.Nm
utility has to be terminated with