Re: [TLS] More clarity on resumption and session hash

[David Benjamin <[e-mail]>]

On Fri, May 29, 2015 at 2:14 PM David Benjamin <[e-mail]>
wrote:

> On Fri, May 29, 2015 at 2:09 PM Salz, Rich <[e-mail]> wrote:
>
>> > I poked a bit more and I was mistaken about OpenSSL's d2i_SSL_SESSION
>> behavior: Although it does ignore the structure version, it will fail the
>> parse if it sees elements at the end of the structure it doesn't understand.
>>
>> We'd like to change this.  Allow applications to append their own data to
>> a session is very useful.
>>
>> (We is probably my employer submitting patches to OpenSSL which someone
>> on the dev team like me will review.()
>>
>
> This is somewhat tangential, but this seems a bad idea. Having
> applications squat the same namespace of tag numbers as OpenSSL will break
> when OpenSSL internally adds more fields to the end. If you want
> applications-specific data in the session, either the application should
> serialize and deserialize a wrapper structure that happens to contain a
> serialized SSL_SESSION, or explicitly introduce an OCTET STRING hole to
> stuff opaque application-specific data into.
>

(Not to mention the obvious bad consequences, such as the one discussed in
this thread, if an older OpenSSL happily parses an SSL_SESSION with fields
it doesn't understand from a new OpenSSL.)

David