Re: [dtn-interest] Comments on RFC 5050

[Michael Noisternig <[e-mail]>]

Am 24.06.2013 20:48, schrieb Amy Alford:
>
>         4.7. Dictionary Revision:
>         "Any [unreferenced] strings (scheme names and SSPs) in a bundle's
>         dictionary ... may be removed from the
>               dictionary at the time the bundle is forwarded."
>         ...subject to that authenticity is not lost.
>
>            * Yes, but that’s implicit.  Strings that are needed in order to
>
>              verify authenticity are referenced, so they can’t be
>         removed from
>              the dictionary.
>
>
>     Could it happen that some extension block be processed, removed and
>     its EID references removed by some intermediate node on a security
>     path? I am still trying to figure out what is possible with the
>     current security spec.
>
> Yes.  Talking through a specific example is probably the best way to
> clarify this.  Suppose we have a very boring network of:
>
> Node A --- Node B --- Node C --- Node D
>
> Node A sends a bundle to node D, with a PIB block with security source
> of node A and security destination of node D.  Node B can add a PIB
> block with security source node B and security destination node C.  When
> node C removes the PIB block destined for node C, it may change the
> dictionary.  But, it would change it back to its state at the time the
> first PIB was calculated.
>
> In any event, PIBs don't cover the dictionary (only the specific entries
> being used in the blocks covered by the PIB).  BABs cover the
> dictionary, but are always the last security block added and first
> removed.  The only time dictionary changes would invalidate a BAB are if
> the dictionary was modified at a non security aware node.  However, if
> the dictionary was being modified because of a change to an extension
> block, that would invalidate the BAB anyway.  The only dictionary
> changes that would invalidate a PIB are gratuitous changes to case
> insensitive fields.
>
> - Amy

Ah, I missed that the PIB does not protect the complete dictionary. 
Otherwise, I was more thinking about partially overlapping paths. 
Something like where Node A adds some (non-security) extension block 
together with some new dictionary entry, both to be processed and 
removed by Node C, whereas Node B adds a PIB to be consumed by Node D.

But what about dictionary entries by non-primary blocks? How are they 
being protected end-to-end?

Michael