Re: [BEHAVE] WGLC: draft-ietf-behave-v4v6-bih-03.txt

> -----Original Message-----
> From: behave-bounces@ietf.org [mailto:behave-bounces@ietf.org] On
> Behalf Of teemu.savolainen@nokia.com
> Sent: Monday, March 21, 2011 4:54 AM
> To: cb.list6@gmail.com; dthaler@microsoft.com
> Cc: draft-ietf-behave-v4v6-bih@tools.ietf.org; behave@ietf.org
> Subject: Re: [BEHAVE] WGLC: draft-ietf-behave-v4v6-bih-03.txt
> 
> IPv4 literals can be used only after a mapping has been created between
> IPv4->IPv6.
> 
> So maybe clarify instead (in the appendix section):
> --
> BIH MAY intercept inet_ntoa() and inet_addr() and use the address
> mapper for those.  Doing that enables BIH to support literal IP
> addresses. However, IPv4 address literals can only be used after a
> mapping entry between the IPv4 address and corresponding IPv6 address
> has been created.

That text can't be an accurate description of the BIH deployed by China
Mobile, which works with Skype.  Skype doesn't do DNS queries for
its DHT or for its peers.

So, I believe the "DNS must be used first" restriction doesn't exist.

-d

> --
> Btw this inet_ntoa/inet_addr text comes from the original BIA, and I
> have to admit I'm not sure what was the intent of the original authors.
> For me it seems those functions just do conversion between IPv4 ASCII
> and IPv4 binary formats (and hence would not necessarily have to be
> here)... but if the "BIA" would allow IPv6 string formats, then the
> inet_addr could maybe convert IPv6 textual presentation into IPv4
> binary format (and create a new mapping/use existing), but I don't see
> how inet_ntoa could convert IPv4 address into IPv6 textual presentation
> without Address Mapper already having an entry for that mapping.
> 
> Best regards,
> 
> 	Teemu
> 
> > -----Original Message-----
> > From: behave-bounces@ietf.org [mailto:behave-bounces@ietf.org] On
> > Behalf Of ext Cameron Byrne
> > Sent: 18. maaliskuuta 2011 20:12
> > To: Dave Thaler
> > Cc: draft-ietf-behave-v4v6-bih@tools.ietf.org; behave@ietf.org
> > Subject: Re: [BEHAVE] WGLC: draft-ietf-behave-v4v6-bih-03.txt
> >
> > the draft states:
> >
> > "The supported class of applications includes those that use DNS for
> >    IP address resolution and that do not embed IP address literals in
> >    protocol payloads."
> >
> > I believe this wording is too strong and makes it sound like BIH
> > cannot support IPv4 literals, when later in the draft it is stated:
> >
> > "BIH MAY intercept inet_ntoa() and inet_addr() and use the address
> >    mapper for those.  Doing that enables BIH to support literal IP
> >    addresses."
> >
> > I suggest this wording to make it clear that IPv4 literals may be
> > supported but embedded addresses with application signalling will
> pose
> >  a challenge
> >
> > "The supported class of applications includes those that use DNS for
> >    IP address resolution.  Direct references of outbound connections
> > to IPv4 literals are also may be  support (ex: http://192.0.2.1)."
> > _______________________________________________
> > Behave mailing list
> > Behave@ietf.org
> > https://www.ietf.org/mailman/listinfo/behave
> _______________________________________________
> Behave mailing list
> Behave@ietf.org
> https://www.ietf.org/mailman/listinfo/behave