[cryptography] if MitM via sub-CA is going on, need a name-and-shame catalog (Re: really sub-CAs for MitM deep packet inspectors?)

Peter Gutmann pgut001 at cs.auckland.ac.nz
Fri Dec 2 08:51:25 EST 2011


Adam Back <adam at cypherspace.org> writes:

>I wonder what that even means.  *.com issued by a sub-CA?  that private key
>is a massive risk if so!  I wonder if a *.com is even valid according to
>browsers.  Or * that would be funny.

No idea, but remember that it's not "general-purpose browsers", it's 
"cellphone browsers" that historically have been crufty little custom apps 
with who knows what behaviour.  Also, the phone's entire worldview is what the 
cell site it's connected to wants it to be.  For example if the telco wants to 
reactivate an expired cert they can just make it be 2006 again.  Or block 
access to CRLs (do mobile browsers even check these?).  Or return '3' in 
response to an OSCP query (assuming mobile browsers do OCSP).

Peter.



More information about the cryptography mailing list