On Wednesday 05 June 2002 23:01, Steven Peck wrote:
> I believe this is it.
<snip>
> In brief, it appears to be a way to establish secure end to end
> communications across NAT and the Internet specificcaly using the
> UPnP standard proposed by Intel.

Though SSH doesn't come out and say this, they are basically the
same idea. NAT causes problems with multiple clients doing the 
*same* thing at the same time. Say like multiple IPSec connections
on port 500 leaving the NAT'ed Gateway. What is proposed here is
a Nat-D type added to the approved header method (tunnel and 
transports are the current standard types). The Nat-D header
would indicate the presence of a second added header that 
includes the port number used by the machine requesting the 
service (IPSec for instance). With this NAT'ed port information
added to the packet payload, the gateway(s) will be able to 
indentify and decode the second header and send it to the 
exact machine that requested the information (identified by the
port the connection was initialized on). 

Although this may not be the best method proposed to deal with
NAT, this is a very easy method to implement and will work on
all NAT and Proxy machines that will support identification and 
routing suggested in the docs. In special cases such as the iSCSI
network storage devices, this can be built directly into the device
driver eliminating the need for encryption by a processor because
it is "built" into the device (driver) itself.

What advantage it would give to us at this time would amount to 
faster thoroughput times and automatic resetting of dropped 
tunnels, assuming that FreeS/WAN supports the changes in any
case.
-- 

~Lynn Avants
aka Guitarlynn

guitarlynn at users.sourceforge.net
http://leaf.sourceforge.net

If linux isn't the answer, you've probably got the wrong question!

_______________________________________________________________

Don't miss the 2002 Sprint PCS Application Developer's Conference
August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm


_______________________________________________
Leaf-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/leaf-devel

Reply via email to