On Thu, 11 May 2000, Theodore Hong wrote:

> You might say falling over still sucks for the epicenter.  But this sort of
> thing can happen under the current system as well -- suppose someone
> misspells a key on slashdot.  Then the epicenter for the misspelling will
> get hit by requests it can't fend off with data because there is no data
> under that key.

And to an even larger extent, with Ian's system of simply deleting
data when an update expires, if the author has not had time to update
when it expires, all attempts to not only update but also retrieve
that data will end up at epi-center node as well, causing the same
slashdot effect (the only difference is with follow-throughs it is
allowed to follow through, with delete on expire it HAS to).

Also along this line, my post from last night in which I claimed that
follow-throughs could be used for DoS was not true. If one wanted to
create such a DoS, one could simply vary the key by the least
signifigant bit, and flood requests for that (sorted to the same
nodes, but guaranteed to follow through until the htl runs out).

> 
> theo
> 
> 
> _______________________________________________
> Freenet-dev mailing list
> Freenet-dev at lists.sourceforge.net
> http://lists.sourceforge.net/mailman/listinfo/freenet-dev
> 


_______________________________________________
Freenet-dev mailing list
Freenet-dev at lists.sourceforge.net
http://lists.sourceforge.net/mailman/listinfo/freenet-dev

Reply via email to