Re: Five Ways to Screw Up SSL

["Dude VanWinkle" <dudevanwinkle () gmail com>]

On 5/22/06, Brian Dessent <brian () dessent net> wrote:
> Valdis.Kletnieks () vt edu wrote:
> >
> > On Mon, 22 May 2006 12:02:23 EDT, Dude VanWinkle said:
> >
> > > DNS foo to the client, how easy is that? Would you have to get the
> > > upstream DNS server to  cache your bogus entry?
> >
> > You'd be *amazed* how many are still running BIND 4 or 8....
>
> That, or use a browser exploit or mass email to drop a small program on
> the end user's machine that changes the configured values of the DNS
> servers to ones that you run.  You don't even have to leave any traces
> of this (like .exe files that run at startup), it could be a one-time
> configuration change.  The end user would have no clue this had happened
> since the malicious servers would be standard recursive resolvers, so
> their net access remains fully operational...


So just taking over their lmhosts would invalidate the chain o' authority.

I guess you would hijack their machines with a bug that would edit the
local cache, refresh the cache, then report to you about the websites
the victim's machine had visited, and you could request an ssl cert
for those sites.


The only problem I see with this scenario from a freessl perspective
is that they require verification in the form of an email sent to
admin () domain com or from an email sent to the admin from the upstream
DNS provider. This would be a little tricky to get around as you would
have to munge freessl's DNS records.

-JP

Gotcha, thanks for the lesson

-JP

_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/