IPv4-mapped IPv6 addresses [no need to read this]

[Julio Baixauli]

Li Hong wrote:
> 
> ...........
>
> I am not sure Linux box, I am using FreeBSD 4.1 as testing bed.
> 
> In my testing, the following testing cases will using  IPv4-mapped IPv6 address:
> 
> 1) IPv4 client communicates with IPv6 server, the server will use IPv4-mapped IPv6 address.
> 
> 2) IPv6 client communicates with IPv4 server, the client will use IPv4-mapped IPv6 address
> 
> 3) IPv6 client communicates with IPv6 server, the client and server will both accept IPv4-mapped IPv6 address.
> 
> Both 1,2 will use IPv4 datagram, 3 will use IPv6 datagram.


	Thank you very much!
	
	This mail won't add any more questions to the list. I only write this
mail to thank everybody that has spent their time in this questions, and
to explain my experience with SIIT algorithm, if anybody are interested.
Now I explain my solution, a temporaly solution until the gurus of IPng
clarify the question about IPv4-mapped addresses.
	So, thank you very much again if you don't want to read more about this
and excuse me for my poor english.

....

	I've been seeing draft-itojun-ipv6-transition-abuse-01.txt, and I see
many problems with IPv4-mapped addresses and SIIT algorithm. This
problems beyond me. There are problems deciding how the kernel handles
IPv4-mapped addresses, problems with the handling of FTP (PORT/PASV &&
EPSV/EPRT), problems of malicious use of these addresses (DoS attacks,
etc...)

	My original intention was test SIIT software (IPv4 to IPv6 headers
translation, and viceversa). The SIIT algorithm says that IPv6 node MUST
use IPv4-mapped address to comunicate to IPv4 node through SIIT
translator. Due to the problems mentioned above, I've decided to make
the test with other types of addresses. I've choosed addresses with the
form ::ffff:ffff:0:0/96 (that looks like ::ffff:0:0/96, IPv4-mapped).
This addresses has been choosed because don't cange the checksum of tcp
headers (this is mentioned in SIIT with IPv4-translatable addresses
too).
	this addresses are handled normaly for the kernel (native IPv6, one
problem less). And don't have the problems of itojun-abuse (bacause the
kernel don't translate them to IPv4)

	At this point, I can test my SIIT software. It seems that work. I can
GET (http) a HTML document of IPv4 node from IPv6 node, translating the
headers in a third node with the SIIT software. I only have made this
test. More tests are comming.

	Again, thank you very much for the help you give me.

	Bye!
-- 
********************************************

	Julio Baixauli Garreta
	baixauli@mat.upc.es

********************************************