Date: Thu, 3 Dec 1998 19:10:01 +0100 From: Guido van Rooij <guido@gvr.org> To: "Jordan K. Hubbard" <jkh@zippy.cdrom.com>, Andreas Klemm <andreas@klemm.gtn.com> Cc: Eivind Eklund <eivind@yes.no>, hackers@FreeBSD.ORG Subject: Re: Can we just come to a decision on IPv6 and IPSec? Message-ID: <19981203191001.A28037@gvr.org> In-Reply-To: <8701.912631921@zippy.cdrom.com>; from Jordan K. Hubbard on Wed, Dec 02, 1998 at 12:52:01PM -0800 References: <19981202172824.A23747@klemm.gtn.com> <8701.912631921@zippy.cdrom.com>
next in thread | previous in thread | raw e-mail | index | archive | help
First, Im no IPV6 expert. Having said that: Today I installed the WIDE stack on my system and it seems that their implementation cannot automatically handle mapped addresses. As an end user on a IPV6 system, when I do telnet foo.bar.org, the resolver will try if there exist an AAAA record for that host and talk IPV6 directly. When no such record is available, it will return an IPV6 mapped address (if I recall the terminology right). In that case, the address (again, if my memory serves me right) ::ffff:a.b.c.d is returned and the kernel can automatically decide that thta is a mapped address and put the packets to the ipv4 stack on the system (with address a.b.c.d) all transparent to the end user. It seems the WIDE stack can not do this (at least not now). They have something called 'faith' which supposedly handles this from userland but I think this should be handled in the kernel. I think handling this transparently is a key issue in the ipv4->v6 migrattion. Does anyone know: 1) If WIDE will deal with this in a later stadium? 2) What the INRIA stack does? -Guido To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19981203191001.A28037>