Re: hop-count header

Duane Wessels <wessels@nlanr.net> writes:

> michael@metal.iinet.net.au writes:
>
> >That's why you increment. Because then the local cache controls
> >policy, not the remote one.
>
> Oh. I thought the problem was that I might trust *my* parent,
> but not necessarily my parent's parent, etc.

That's right. You decide what you'll trust, and thus you set the max
hop-count you'll accept. The issue is things like.

                A
                |
                B
               / \
              C D
[ read, B parents thru A, C parents thru B etc ]

C trusts A and B, but D only trusts B. If you increment the hop count,
then C and D can both get what you they want. If you decrement the
hopcount, it's actually A that sets the policy, which is (IMHO at
least :) the wrong way around.

Michael.
Received on Tue Jul 29 2003 - 13:15:44 MDT