> Thank you for the clarification, Tim.
> That is exactly what I wanted to say. :)
>
> By the way, as regards recent Bind 9, birthday attack is much more
> difficult to conduct because even if the attacker sends multiple
> simultaneous recursive queries, Bind 9 aggregates these queries.
Aggregating queries would definitely help if you assume the attacker can
make recursive queries.
However, it was my understanding (which could be completely wrong) that
BIND 9 reuses sockets for multiple queries, unlike previous versions,
and this makes spoofed attacks easier in another respect. (Of course
this all has nothing to do with the Windows-specific flaw.)
> In addition, there is a patch written by Jinmei-san for Bind 9.4.0
> (current release) to randomize source ports.
>
> http://www.jinmei.org/bind-9.4.0-portpool.patch
> http://member.wide.ad.jp/tr/wide-tr-dns-bind9-portpool-01.txt
> (technical report from WIDE project in Japanese)
That's good, that at least someone is trying to do this in BIND.
> Thank you for the clarification, Tim.
> That is exactly what I wanted to say. :)
>
> By the way, as regards recent Bind 9, birthday attack is much more
> difficult to conduct because even if the attacker sends multiple
> simultaneous recursive queries, Bind 9 aggregates these queries.
Aggregating queries would definitely help if you assume the attacker can
make recursive queries.
However, it was my understanding (which could be completely wrong) that
BIND 9 reuses sockets for multiple queries, unlike previous versions,
and this makes spoofed attacks easier in another respect. (Of course
this all has nothing to do with the Windows-specific flaw.)
> In addition, there is a patch written by Jinmei-san for Bind 9.4.0
> (current release) to randomize source ports.
>
> http://www.jinmei.org/bind-9.4.0-portpool.patch
> http://member.wide.ad.jp/tr/wide-tr-dns-bind9-portpool-01.txt
> (technical report from WIDE project in Japanese)
That's good, that at least someone is trying to do this in BIND.
thanks for the info,
tim
[ reply ]