From: Carsten Bormann <[email protected]>
Date: Wed, 10 Jun 2009 21:31:40 +0200
On Jun 10 2009, at 20:39, Richard Kelsey wrote:
> By the way, if two devices with the same EUI64 are powered
> up or reboot at the same time, it looks to me as if the
> conflict might never be detected. Assuming that the Edge
> Router gives them the same registration lifetimes, the two
> TIDs will move forward in leap frog fashion, never getting
> more than one removed from each other.
I thought that for a while, too.
However, a registration with the same TID is defined to cause a
conflict.
So if the nodes don't manage to exactly synchronously alternate in
losing packets, eventually the conflict will be detected.
In section 7.7, if TID1 is equal to TID2 are the two
consistent? In other words does zero count as a small
increment/decrement?
If receiving the same TID twice is treated as an OII
collision, what mechanisms are there for detecting
duplicate packets?
Now that I have just read 7.7 again, at the very end it
says:
If the new value is not consistent with a recent value
saved in the whiteboard entry then it is rejected as a
collision.
>From context, I think "rejected as a collision" means that
the TID does not indicate a collision and thus is accepted,
but the wording could be clearer.
Unless we require hosts to be awake enough to defend their addresses,
there is no discernable difference between a reboot and a conflict of
a new node with a sleeping node. So it will be hard to enable
detection immediately after switching on the colliding node.
That is certainly true if hosts have no available
non-volatile storage. Even if they do have such storage, it
isn't clear how much it would help to make use of it.
-Richard Kelsey
_______________________________________________
6lowpan mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/6lowpan
