Support for EC crypto?

[Cantor, Scott]

On 9/13/15, 6:15 PM, "users on behalf of Stefan Santesson" <users-bounces at shibboleth.net on behalf of stefan at aaa-sec.com> wrote:

>Let me immediately expand the question also to the use of RSASSA-PSS (http://www.w3.org/2007/05/xmldsig-more#rsa-pss) for signatures.

I am not familiar with that, so I would assume that it isn't supported by Shibboleth or any of the libraries we have to use. It seems to be in OpenSSL at least.

>A main reason for my question is that I currently review the new EU interop specifications for corss-border SAML. It seems to be anything but main-stream.
>It requires RSASSA-PSS 3072 for signatures on SAML messages and AES-128-gcm or AES-256-gcm for encrypting XML objects such as assertions.
>It also allows EC
>
>I have a feeling that this will cause major problems and use of custom code that will introduce far more security problems than the algoritms ans modes they try to ban.

I think you are correct. I can't really fault requiring AES-GCM, since it is the only secure algorithm in XML Encryption now, but wishing doesn't make it work.

I don't think us testing GCM or EC and getting them working properly if there's something left to do is a bad thing, but it takes a lot of time from pretty much either Brent or myself. If somebody wants to offer, I'll certainly help.

But I don't know anything about RSA-PSS, no.

-- Scott