freebsd4.11 patch for nfs over tcp
Peter Edwards
peadar.edwards at gmail.com
Tue Jul 25 14:53:17 UTC 2006
On 7/25/06, Jim Rees <rees at umich.edu> wrote:
> I see by inspecting the code that you are right and the web site I consulted
> was wrong. Don't believe everything you read on the web.
>
:-)
I always point people at "Rethinking the TCP Nagle Algorithm" (Mogul &
Minshall) when this comes up. It's a nice readable paper on the issues
and some propsed solutions.
http://www.acm.org/sigs/sigcomm/ccr/archive/2001/jan01/ccr-200101-mogul.pdf
> I also see that Nagle himself suggests not using his algorithm with delayed
> acks. It can work, but all the timers have to be just right.
>
> Anyway this bug was fixed a while ago, and I don't think there will be
> another 4.x kernel, so there's not much to be done about this.
>
Absolutely. just setting things straight "for the record."
More information about the freebsd-fs
mailing list