Date: Mon, 28 Jul 2003 14:27:07 +0400 From: Gleb Smirnoff <glebius@cell.sick.ru> To: Brett Glass <brett@lariat.org> Cc: net@freebsd.org Subject: Re: NAT and PPTP Message-ID: <20030728102707.GB65369@cell.sick.ru> In-Reply-To: <4.3.2.7.2.20030723192331.02c9bbd0@localhost> References: <20030723213028.GB48101@sunbay.com> <4.3.2.7.2.20030723192331.02c9bbd0@localhost>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Jul 23, 2003 at 07:27:38PM -0600, Brett Glass wrote: B> I haven't gotten any hopes up, but it would be nice. It seems as B> if the only alternatives are to un-GNU PoPToP (which requires B> a clean room team; possible but not easy) or to create a FreeBSD B> pptpd that is analogous to pppoed. This would use your work, Archie, IMHO, this is not good idea. Currently I'm running two different types of access points: 1) PPPoE concentrators: pppoed + ppp 2) PPTP server: mpd with huge mpd.conf and mpd.links As I remember, Brett said that mpd allocates a number of nodes for each connection in kernel memory. That's right. But in case of pppoed+ppp or imaginary pptpd+ppp you will have a user-level process and ng_socket for each connection. Not shure that it will take less memory. But it will do a lot of context switching. On my own experience it looks like PPTP (no comperssion, no encryption) access point with mpd is more robust than PPPoE one with pppoed+ppp. Currently I'm planning to look into Alexandr Motin's patches giving PPPoE server support for mpd. -- Totus tuus, Glebius. GLEBIUS-RIPN GLEB-RIPE
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030728102707.GB65369>