Ian wrote :
> Done - the exact threshold can be configured in the .freenetrc file.
just an idea,
What if you up this threshold to say half the datsore size limit for
inserts, that way you automaticly have countered the attack where you try to
fill up the datastore with targeted inserts to make som
It would seem that the "weight" function should be proportional to the
size of the file *multiplied* by the time since last request. This seems
intuitive since the resources consumed are proportional to both the size and
the time the item is alive.
But an efficient datastructure to manage this is
On Mon, 07 Aug 2000, Ian Clarke wrote:
<>
> Ok, how is this for a solution to this exploit: Rather than using the
> position in the datastore found by the method I describe as the insert
> point for the data, it is used as a lower-bound for the insert point,
> ie. the data is inserted at a random
> > Ok, how is this for a solution to this exploit: Rather than using the
> > position in the datastore found by the method I describe as the insert
> > point for the data, it is used as a lower-bound for the insert point,
> > ie. the data is inserted at a random position above this point.
>
> T
> > c) You'd need to make sure searchData() does the same thing (if you want it
> > too).
> H, I assumed that searchData() would use the put() method, but I will
> check this.
Yep, its cool - searchData() uses the put() method, so the mod affects
it already.
Ian.
___
> > a) So if I take a key value and insert exactly one byte of data for it, that
> > data will never expire? What if I flood Freenet with thousands of one byte
> > inserts? Will they be blocking datastores (which are limited by number of
> > entries as well) and never be cleared?
> This may be an
> a) So if I take a key value and insert exactly one byte of data for it, that
> data will never expire? What if I flood Freenet with thousands of one byte
> inserts? Will they be blocking datastores (which are limited by number of
> entries as well) and never be cleared?
This may be an issue, but
a) So if I take a key value and insert exactly one byte of data for it, that
data will never expire? What if I flood Freenet with thousands of one byte
inserts? Will they be blocking datastores (which are limited by number of
entries as well) and never be cleared?
b) Isn't this a little risky do r
I have now modified the datastore so that it takes size into account by
creating a new insert() method in CyclicArray, and modifying the put() method
in StandardDataStore.
This piece of javadoc describes the new functionality of put():
"""
Inserts a DataStoreItem into the datastore. The new item
