Date: Fri, 25 Jul 2014 19:11:34 -0400 (EDT) From: Garrett Wollman <wollman@freebsd.org> To: rmacklem@uoguelph.ca Cc: stable@freebsd.org Subject: Re: nfsd server cache flooded, try to increase nfsrc_floodlevel Message-ID: <201407252311.s6PNBYid087959@hergotha.csail.mit.edu> References: <53D23B57.8020208@omnilan.de> <666117227.3741955.1406328843281.JavaMail.root@uoguelph.ca>
next in thread | previous in thread | raw e-mail | index | archive | help
In article <666117227.3741955.1406328843281.JavaMail.root@uoguelph.ca>, Rick Macklem writes: >The problem with setting vfs.nfsd.tcphighwater too large is that an overloaded >server can run out of kernel address space/memory it can use for malloc() and >then the whole machine wedges. (I can only test to see what is safe for a >256Mbyte i386.) Rick, The FreeBSD Foundation maintains significant resources in the various FreeBSD clusters for the purpose of testing FreeBSD on a variety of hardware platforms and scales. Why don't you ask them for access to some of these systems so you can run tests of your own? I don't think it would be hard to come up with a test plan that would cover at least the most common cases. -GAWollman
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201407252311.s6PNBYid087959>