UTA members,

With the Heartbleed exploit all over the news lately, it was pointed out to me 
that web *client* applications could be vulnerable to this in that if they 
visit a site operated by a malicious server, the server could exploit the bug 
to capture memory blobs from the client systems.  There is even demonstration 
code out there that shows how this can work:

https://github.com/Lekensteyn/pacemaker

Now, the path to exploit this on the client side might be more difficult 
because:
1. The attacker would have to get the client to visit the malicious site (i.e. 
via phishing or shared links on social media); or
2. The attacker would have to do a MiTM or modify code on an existing 
legitimate site; and
3. The attacker would have to repeatedly be retrieving memory blobs to try to 
see if anything interesting was in there.

Still, the possibility is out there that the client can leak information.  Much 
is being made in the media right now about *servers* updating their use of 
OpenSSL, but I've not seen much yet about clients that link to the OpenSSL 
libraries to make TLS/SSL connections.

I guess my thought is... as we develop documents for application developers 
about how to best use TLS in their apps, are there lessons we can take from all 
of this to add into those documents?     (And I haven't really thought through 
it all myself, yet... more just posing the question.)

Dan

--
Dan York
Senior Content Strategist, Internet Society
[email protected]<mailto:[email protected]>   +1-802-735-1624
Jabber: [email protected]<mailto:[email protected]>
Skype: danyork   http://twitter.com/danyork

http://www.internetsociety.org/deploy360/

_______________________________________________
Uta mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/uta

Reply via email to