Hi, I agree the removal.
We are working for importing IPv6 to our embedded OS. At initial phase, we used v4-mapped address to upgrade server and everything worked fine. But when we needed more detail control, we failed to set IP layer socket option with AF_INET6 socket. So the modification of the listen(::) + listen(0.0.0.0) was needed finally. -Alan -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of sasson, shuki Sent: Tuesday, February 01, 2005 11:14 PM To: Jeroen Massar; Markku Savela Cc: Bob Hinden; ipv6@ietf.org Subject: RE: IPv6 Address Architecture update question Listening for both IPv4 and IPv6 connection is very useful for servers. Examples ftp server, SNMP agent. Servers usually don't care about the source address since they are responding to source no matter if this is an IPv4 address or IPv6 address. The existing interface allows them to do it with the same loop as for IPv4. Thus porting of servers becomes much easier. Connecting clients do need to know about the destination addresses. Still keeping the current interface for servers is a must! Shuki > > From: Jeroen Massar <[EMAIL PROTECTED]> > > > Thus that applications should be using multiple sockets, one for > > IPv4 and one for IPv6 and one for whatever follows. > > I strongly object to this. There are other socket api's which don't > have the Unix inherited drawbacks. For such, the recommendation is > exactly the opposite: the same socket works just fine for IPv4 and > IPv6, and in unifying the application code to work for both, IPv4 > mapped address format is very useful tool. Any documentation on this ? I do know RFC3493, not any others. > There is no reason for application to care at all whether actual > connection is over IPv4 or IPv6. There is, these are different protocols, how else are you going to address which host you are going to connect to? For instance www.kame.net port 80 results in different answers over IPv6 than the one on IPv4... If you are talking about wrappers, then these are most likely wrapping around the above already anyway, thus it matter than? Greets, Jeroen -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jeroen Massar Sent: Tuesday, February 01, 2005 9:46 AM To: Markku Savela Cc: ipv6@ietf.org; Bob Hinden Subject: Re: IPv6 Address Architecture update question -------------------------------------------------------------------- IETF IPv6 working group mailing list ipv6@ietf.org Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 -------------------------------------------------------------------- -------------------------------------------------------------------- IETF IPv6 working group mailing list ipv6@ietf.org Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 -------------------------------------------------------------------- -------------------------------------------------------------------- IETF IPv6 working group mailing list ipv6@ietf.org Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
