Hello,
I recently had a look at how browsers react to DH key exchanges with
bogus modulus values. here's what I found:
http://blog.hboeck.de/archives/841-Diffie-Hellman-and-TLS-with-nonsense-parameters.html
And here is a test (warning: crashes some chrome/chromium versions)
https://dh.tlsfun.de/
This is a hard-coded patch to make OpenSSL clients reject connections
which use DHE handshakes with 1024 bits.
This patch has no compile-time or runtime configurability. If the
project wants something more nuanced, we need discussion about what
the right form(s) of configurability should be.
On 03/13/2014 06:33 AM, Hanno Böck wrote:
I recently had a look at how browsers react to DH key exchanges with
bogus modulus values. here's what I found:
http://blog.hboeck.de/archives/841-Diffie-Hellman-and-TLS-with-nonsense-parameters.html
And here is a test (warning: crashes some
On 03/13/2014 04:05 PM, Kurt Roeckx wrote:
On Thu, Mar 13, 2014 at 03:13:01PM -0400, Daniel Kahn Gillmor wrote:
In theory, users of OpenSSL as a TLS client are already able to query
the size of the DH key exchange for any given connection, and can choose
to terminate it if they object to the
On Thu, Mar 13, 2014, Kurt Roeckx wrote:
On Thu, Mar 13, 2014 at 03:13:01PM -0400, Daniel Kahn Gillmor wrote:
In theory, users of OpenSSL as a TLS client are already able to query
the size of the DH key exchange for any given connection, and can choose
to terminate it if they object to the
On Thu Mar 13 20:12:38 2014, d...@fifthhorseman.net wrote:
This is a hard-coded patch to make OpenSSL clients reject connections
which use DHE handshakes with 1024 bits.
I should've commented on this before, sorry. I'm currently working on a
framework where several security parameters can be
On 03/13/2014 05:52 PM, Stephen Henson via RT wrote:
I should've commented on this before, sorry. I'm currently working on a
framework where several security parameters can be configured at both compile
time and runtime, including DH parameter sizes. It's still under development
at
present
