Date: Sat, 30 Jul 2005 19:00:19 +0930 From: "Daniel O'Connor" <doconnor@gsoft.com.au> To: Julian Elischer <julian@elischer.org> Cc: freebsd-net@freebsd.org Subject: Re: AltQ + ng_iface Message-ID: <200507301900.32660.doconnor@gsoft.com.au> In-Reply-To: <42E99CFD.6070803@elischer.org> References: <200507290834.10268.doconnor@gsoft.com.au> <42E99868.1080306@mac.com> <42E99CFD.6070803@elischer.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--nextPart2598972.cpXo4H64aI Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline On Friday 29 July 2005 12:35, Julian Elischer wrote: > I do this to great effect.. > consider: > two sites connected by links in which teh bottleneck is 200KB/sec (1 E1?) > when a lot of data is flowing from 1 to 2 then data from 2 to 1 is also > slowed > down because the acks have to go through the queues on ingress side of the > bottleneck router. > > I add a dummynet entry on 1, limiting output to 190KB/sec, so that the > queue is in dummynet and not the intermediate router, and then allow small > ack packets > to bypass that queue. As a result the data from 2 to 1 also flows at > near capacity, > and with a much lower latency. SInce data flows tend to be large packets, > I sometimes actually prioitise ALL small packets allowing interactive > stuff to > bypass ftps etc. and sometimes I do it on both ends. I tried to update my ipfw setup, but I could not manage to get ipfw + natd = to=20 work with stateful rules :( Although since natd runs in userland I may as well switch back to ppp(8) to= =20 save extra kernel/userland transitions. I was hoping to be able to use pf += =20 mpd since that puts all of the packet processing into the kernel. I did try pf + ipfw + dummynet (eww) but it appears that when the packets g= et=20 reinjected back into the system after the pipes they go through pf again=20 which doesn't like them (not 100% sure why.. maybe they don't match a state= =20 properly any more?) I'll try my hand at the ng_iface ALTQ patch. =2D-=20 Daniel O'Connor software and network engineer for Genesis Software - http://www.gsoft.com.au "The nice thing about standards is that there are so many of them to choose from." -- Andrew Tanenbaum GPG Fingerprint - 5596 B766 97C0 0E94 4347 295E E593 DC20 7B3F CE8C --nextPart2598972.cpXo4H64aI Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.1 (FreeBSD) iD8DBQBC60i45ZPcIHs/zowRAjuNAJwOb4kntBP0Ah0+fBhmbJ0nlpcpwgCfQL0W y2B6g2LZAn0qRXRWJRp1v0U= =kTAS -----END PGP SIGNATURE----- --nextPart2598972.cpXo4H64aI--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200507301900.32660.doconnor>