Date: Wed, 30 Jul 2003 11:56:21 +0200 From: =?iso-8859-1?Q?Sten_Daniel_S=F8rsdal?= <sten.daniel.sorsdal@wan.no> To: "Christophe Prevotaux" <c.prevotaux@hexanet.fr>, "Brett Glass" <brett@lariat.org> Cc: net@freebsd.org Subject: RE: NAT and PPTP Message-ID: <0AF1BBDF1218F14E9B4CCE414744E70F07DEFA@exchange.wanglobal.net>
next in thread | raw e-mail | index | archive | help
> My own purpose for using this is securing a bit more=20 > 802.11(whatever) in a > large WISP setup. One of my question is how many pptp or=20 > pppoe sessions=20 > can be handled by one FreeBSD box knowing each pptp or pppoe=20 > sessions have > to be shaped traffic wise symetrically or asymetrically.=20 depends on the box, the shaping is very efficient and is in kernel. so it's primarily the pptp and pppoe sessions that will demand = resources. it's almost impossible to answer. if i said at least 30 on a celery 1ghz, you might/might not be happy = with that. im sure you could run twice that on the same hardware (given it's = good hardware and is not the cheapest sh*t you could find). > So having the ability to shape inbound bandwidth and outbound=20 > bandwidth directly > inside the pptpd and pppoe thru radius and directly (for some=20 > cases) thru ppp.conf > would be really nice (it would require having a special=20 > dictionary for radius (I think)) > I don't know if this is achievable without too much hassle in=20 > the current PPP (PPPOE) > code and if it is at all possible in a PPTP environment? >=20 i use a shell script called from ppp.linkup/ppp.linkdown under the = appropriate label (radius supplies label as "Filter-Id"). Need it be simpler? - Sten
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?0AF1BBDF1218F14E9B4CCE414744E70F07DEFA>