Hello all!
I'm developing a micro-mobility protocol in an IPv6 network for my
final degree project but I'm having some problems using IPv6.

Imagine the following, simple, situation:
I'm connecting only two machines, a gateway and a pc that will work as
an access point using hostap. I have assigned global IPv6 addresses
(2000::etc/3) to simplify things (the protocol v4 is using global ipv4
for the machines) and my network is an island isolated from the
internet.

I've configured the routes to be point-to-point from the GW to the AP
and vice-versa.
When I try to ping the AP from the GW, the src addr of the packets
transmited from the GW is the global IPv6 addr that I've assigned to
the GW (so far so good), but when I try to ping the GW from the AP,
the src addr of the packets transmited from the AP is the link local
IPv6 addr autoconfigured by the AP... why is that? shouldn't the src
addr be the global addr ? they behave differently, the GW and the
AP...in either situations, the src addr should be OR the link local,
or the global addr, but the should be the same type of addr in both
situation.

The second problem is when I turn on ipv6 forwarding using the echo 1
> /proc/.../ipv6/conf/all/forwarding.
When I do this, if I ping the AP from the GW, the echo request gets to
the AP but the AP never answers with an echo reply..
If I try to ping the GW from the AP I get the message "connect:
network is unreachable"

It sounds weird that turning on ipv6 forwarding "destroys" the
*simple* ping point-to-point between AP and GW...

If somebody could help with this two issues I would be very gratefull :)
I'm using debian with a 2.6.8 kernel.
Thanks,



~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Pedro Tome'  [EMAIL PROTECTED]
INESC-id Grupo de Computa��o Natural
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

---------------------------------------------------------------------
The IPv6 Users Mailing List
Unsubscribe by sending "unsubscribe users" to [EMAIL PROTECTED]

Reply via email to