Date: Sat, 28 Mar 1998 11:43:25 -0500 From: dennis <dennis@etinc.com> To: Dan Janowski <danj@3skel.com>, freebsd-hackers@FreeBSD.ORG Subject: Re: Network throttle... Message-ID: <3.0.32.19980328114322.00ff9ea0@etinc.com>
next in thread | raw e-mail | index | archive | help
At 10:35 AM 3/28/98 -0500, Dan Janowski wrote: >I am looking for a way to create a lower bandwidth restriction on an >interface. I have an ethernet interface that I want to restrict to a >maximum >of T1 throughput. > >1. Is there a way to do this (easy or kernel hacking) We have a commercial add-on product that does this http://www.etinc.com/bwmgr.htm > >What happens, i.e. the mechanics, when there is a higher >capacity link feeding into a lower capacity link. Where does all >the data go? I know it is buffered to an extent, but that has limits. >Do the packets get dropped? I seem to remember an ICMP type that >indicates transmission of too much data and for the sender to cut back >or something. Our product "simulates" the functioning of IP packets when they reach a slower link. Data is only dumped when the hold time is greater that a threshold (as TCP will retransmit anyway). In normal operation (ie without substantial overload like video feeds or flood pinging) TCP is self-throttling by "slowing the acks". Source quenching is not necessary, unreliable and generally creates a mess. You can limit the entire interface, limit individual hosts by IP address, entire networks or limit data types (like cu-seeme traffic). Dennis > >All thanks, > >Dan > > >-- >danj@3skel.com >Dan Janowski >Triskelion Systems, Inc. >Bronx, NY > > > > >To Unsubscribe: send mail to majordomo@FreeBSD.org >with "unsubscribe freebsd-hackers" in the body of the message > > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3.0.32.19980328114322.00ff9ea0>