On Mon, Oct 28, 2002 at 07:58:11PM -0500, Greg Wooledge wrote:
> Matthew Toseland (toad at amphibian.dyndns.org) wrote:
> 
> > BUT this will then count under the bandwidth limiter, i.e. be very very
> > slow.
> 
> I don't use the bandwidth limiters. ;-)
> 
> > Do we want to count certain IP ranges as local and not limit them,
> > or do we want to never limit mainport connections, or what? Which is the
> > best solution?
> 
> My thought is that anything talking to mainport (or nodestatus)
> should never be limited artificially.  This may break down in the
> pathological case where people use allowedHosts=* (e.g. for a
> public Freenet gateway).  But I think *most* people are only going
> to allow LAN and loopback connections to mainport.
> 
> I'd urge people who are serious about bandwidth shaping to look into
> their operating system's capabilities instead of relying on the
> applications to do the right thing.  The OS is usually far more
> reliable in this area.
The OS is operating at the wrong level. TCP cannot be shaped without
cooperation from the other end at a higher level than TCP.
> 
> -- 
> Greg Wooledge                  |   "Truth belongs to everybody."
> greg at wooledge.org              |    - The Red Hot Chili Peppers
> http://wooledge.org/~greg/     |



-- 
Matthew Toseland
toad at amphibian.dyndns.org
amphibian at users.sourceforge.net
Freenet/Coldstore open source hacker.
Employed full time by Freenet Project Inc. from 11/9/02 to 11/11/02.
http://freenetproject.org/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: 
<https://emu.freenetproject.org/pipermail/devl/attachments/20021029/3eaed8da/attachment.pgp>

Reply via email to