Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!newshost.telstra.net!act.news.telstra.net!psgrain!news.uoregon.edu!news.bc.net!news.mindlink.net!van-bc!ddsw1!news.mcs.net!not-for-mail From: les@MCS.COM (Leslie Mikesell) Newsgroups: comp.unix.bsd.freebsd.misc Subject: Re: ip-masq Date: 28 Feb 1996 11:51:55 -0600 Organization: /usr/lib/news/organi[sz]ation Lines: 32 Message-ID: <4h24nr$plo@Mercury.mcs.com> References: <4gr0kb$d4@uriah.heep.sax.de> <3132E3F1.446B9B3D@kampai.euronet.nl> <4gv0s5$1nq@agate.berkeley.edu> <4h1r4j$ocl@oz.plymouth.edu> NNTP-Posting-Host: mercury.mcs.com In article <4h1r4j$ocl@oz.plymouth.edu>, Ted Wisniewski <ted@oz.plymouth.edu> wrote: >>The source for IP_MASQUERADING information is ftp.eves.com:/pub/masq/ >>Also included at that site is information about what IP >>masquerading is, and how it works. >> >>I guess this is just yet another cool feature that Linux has and >>FreeBSD doesn't. > > Sounds like someone re-wrote slirp and integrated it and called >it a feature. ;-> It would be a feature that many places need, and given that slirp is based on the bsd network code perhaps it wouldn't be too hard to integrate it back. Many places have anywhere from a couple of machines to an office full using private IP numbers and have an internet account with a single 'real' address. Or you have a class C but too many machines. Most of the machines only need outbound connections, so a socks-style gateway would work, but then you have to dig up socks-aware apps for all your machines. What we need is address remapping where you can just point the default router address at the server. Slirp does this nicely but only for serial line connections. We need it for network links as well. Then you can drop in a single machine that acts as an SMTP/POP server (plus ftp/http if you need that) to handle all the inbound connections on the 'real' address, plus acting as the router/gateway for the hidden net. You can get this effect if you have slirp on the other end of your connection, but that doesn't match the common topology. Les Mikesell les@mcs.com