I'd be curious to see the output of netstat -i and ethtool ethX on both
source and destination machines...
Jerry Wilborn
jerrywilborn@gmail.com
On Tue, Jan 25, 2011 at 9:01 PM, Joey Kelly <joey@joeykelly.net> wrote:
> On Tue January 25 2011 8:45 pm, Lee S. Whatley wrote:
> > On Jan 25, 2011, at 8:21 PM, Petri Laihonen wrote:
> > > net.core.rmem_default = 524288
> > > net.core.rmem_max = 524288
> > > net.core.wmem_default = 524288
> > > net.core.wmem_max = 524288
> > > net.ipv4.tcp_wmem = 4096 87380 524288
> > > net.ipv4.tcp_rmem = 4096 87380 524288
> > > net.ipv4.tcp_mem = 524288 524288 524288
> >
> > I'm curious where you got these numbers from, specifically the 524K
> > buffer size number. Most of the things I've read on tcp/ip
> > optimization suggest much larger numbers, like 16 or 32 MB (i.e.
> > 16777216 versus 524288) for your buffer sizes.
>
> I read last week about bufferbloat... setting larger buffers would add to
> the
> effect, no?
>
> --
> Joey Kelly
> Minister of the Gospel and Linux Consultant
> http://joeykelly.net
> 504-239-6550 <tel:+15042396550>
> ___________________
> Nolug mailing list
> nolug@nolug.org
>
___________________
Nolug mailing list
nolug@nolug.org
Received on 01/25/11
This archive was generated by hypermail 2.2.0 : 01/25/11 EST