From owner-freebsd-current Mon Apr 22 23:51:34 1996 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id XAA08730 for current-outgoing; Mon, 22 Apr 1996 23:51:34 -0700 (PDT) Received: from precipice.shockwave.com (precipice.shockwave.com [171.69.108.33]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id XAA08725 for ; Mon, 22 Apr 1996 23:51:31 -0700 (PDT) Received: from localhost.shockwave.com (localhost.shockwave.com [127.0.0.1]) by precipice.shockwave.com (8.7.5/8.7.3) with SMTP id XAA12653; Mon, 22 Apr 1996 23:49:24 -0700 (PDT) Message-Id: <199604230649.XAA12653@precipice.shockwave.com> To: Terry Lambert cc: msmith@atrad.adelaide.edu.au (Michael Smith), current@FreeBSD.org Subject: Re: socks support native in freebsd? In-reply-to: Your message of "Mon, 22 Apr 1996 23:33:29 PDT." <199604230633.XAA18930@phaeton.artisoft.com> Date: Mon, 22 Apr 1996 23:49:23 -0700 From: Paul Traina Sender: owner-current@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk From: Terry Lambert Subject: Re: socks support native in freebsd? By IP tunneling the default route to the socksd that then forwards it to the forwarding host using a static route to the real interface. Local routes can also go to the local linterface statically, by net. Socks *functionality* is the ability to forward packets by proxy through a connection to a proxy agent on the firewall. This has the benefit of *not* implementing the "functionality" of "recompile all socket using programs". > > It is a mistake to rebuild "telnet, ftp, et al" to achieve > > functionality that belongs at the transport layer, not in the > > applications. > > The functionality is correctly implemented in the _library_ functions > that telnet, ftp et. al call. Incorporating Socks proxy support in the > system libraries would instantly Socksify _all_ of the system, including > any ports built, as well as Perl and so forth. > > Given the popularity of firewalls these days, this would be a Big Plus. First, he was talking about implementing it on a per application basis via the makefile hack that is recommended by the socks package. True. Second, this is an atypical network configuration, and the average user should not have to pay for it in their libc. Untrue, however that's what libsocks.so is for. However, the point is moot, since socks5 uses LD_PRELOAD to implement its magic, socks5 can be made a "real" package because there is no need to modify our code to link in the proxy fake-out code. I'm going to back out my changes for socks4, they're irrelevant with socks5.