Date: Wed, 8 Jun 2005 15:20:51 +0300 From: Giorgos Keramidas <keramida@FreeBSD.org> To: Jeremie Le Hen <jeremie@le-hen.org>, freebsd-net@FreeBSD.org Subject: Re: Problems with gif tunnels Message-ID: <20050608122051.GA1650@orion.daedalusnetworks.priv> In-Reply-To: <20050608104053.GK41050@obiwan.tataz.chchile.org> References: <20050607093717.GA76296@wantadilla.lemis.com> <20050607100958.GU41050@obiwan.tataz.chchile.org> <20050607093717.GA76296@wantadilla.lemis.com> <20050607094848.GB16223@stack.nl> <20050607231218.GD64194@wantadilla.lemis.com> <20050608084946.GI41050@obiwan.tataz.chchile.org> <20050608095703.GM64194@wantadilla.lemis.com> <20050608195837.Q65103@Neo-Vortex.net> <20050608104053.GK41050@obiwan.tataz.chchile.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2005-06-08 12:40, Jeremie Le Hen <jeremie@le-hen.org> wrote: > > IIRC, > - Linux uses the ipip module to do IP-over-IP tunnel > - FreeBSD uses the gre(4) interface to do GRE tunnels > - GRE is a Cisco product and means ``Generic Routing > Encapsulation''. I don't know what they mean with the term > "Generic" because I have only seen IP encapsulated tunnel so > far. It's not mandatory to tunnel IP over GRE. Other, non IPv4 protocols can be encapsulated in the GRE payload. > According to the GRE header, I guess GRE is far more powerful > than a simple IP-over-IP encapsulation, and I would be glad if > someone could explain us what are the benefits of this protocol. GRE is not a simple encapsulation of an IP packet as the payload of an IP packet. It also includes fields in the outter IP header that may support, among other things: - Source routing - A crude form of authentication (key & sequence number pair) - Separation of tunelled "flows" without the payload inspection IP-IP tunnels may require
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050608122051.GA1650>