Matthew Toseland wrote:
On Fri, May 19, 2006 at 01:45:26PM -0500, David Sowder (Zothar) wrote:
It seems to me that a node could benefit from knowing that it doesn't
need to do outgoing hole punching.
Some possible benefits:
- The node wouldn't need to send handshakes to disconnected nodes
nearly as often for nodes we haven't heard from in some time period
As long as it knows the other end also is not firewalled.
Are there firewalls that do UDP hole punching based on receiving packets
from outside rather than from inside?
- The node wouldn't need to do ARK requests nearly as often for nodes
we haven't heard from in a while either
Their IP address may have changed; ours may have changed.
If theirs changed, we'll still get handshakes from them. If ours
changed, we'll know it and can reset all of our "heard from in a while"
timers.
- Reduces output bandwidth demand, thus freeing it for other traffic
- This could help https://bugs.freenetproject.org/view.php?id=316
some at least
I'm not convinced it's a problem.
I assume you're talking about not being convinced bug #316 is a
problem; it's probably not a problem so much as an annoyance.
_______________________________________________
Devl mailing list
[email protected]
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
