RE: [p2p-hackers] Strange Behavior...Concerning NATs

Wed, 05 Jul 2006 21:35:27 -0700

That's my call as well. The NAT doesn't do hairpinning of inside->outside traffic addressed to one of its own external addresses.
 
One case I've seen that is exceptionally hard to deal with is two nodes, one behind a non-hairpinning NAT, the second behind a second NAT (a wireless access point in NAT mode, for example) that is behind the first non-hairpinning NAT. There is no way for them to communicate using their public addresses (because the outer NAT doesn't hairpin), nor any way for the second node to discover the address on the "outside" of the second NAT that the first node could talk to it on. This goes from hard to impossible if the two NATs are assigning "inside" adresses from the same private address space (not uncommon).
 
Matthew Kaufman
 

From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of David Barrett
Sent: Wednesday, July 05, 2006 8:19 PM
To: 'theory and practice of decentralized computer networks'
Subject: RE: [p2p-hackers] Strange Behavior...Concerning NATs

Could also be a NAT that doesn’t support hairpinning.

 

-david

 

From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Lemon Obrien
Sent: Wednesday, July 05, 2006 7:32 PM
To: theory and practice of decentralized computer networks
Subject: [p2p-hackers] Strange Behavior...Concerning NATs

 

When two processes are running on the same machine; different port numbers, using their known global ip address, can not talk to each other, yet be able to find and communicate with all other peers.

 

Does anyone recognize the pattern and understand what might be happening?

 

thanks

_______________________________________________
p2p-hackers mailing list
[email protected]
http://lists.zooko.com/mailman/listinfo/p2p-hackers

Reply via email to