On 10/9/2013 11:32 AM, Robert Guerra wrote: > From the news i've read... a couple of questions for the pfsense developers > come to mind: > > 1. Random Number generation > - NSA is reported to have weakened several random number generators and/or > introduced vulnerabilities. > - What is used in PFsense?
We use the RNG from FreeBSD, which can be assisted by hardware, assuming you trust the hardware. http://en.wikipedia.org/wiki//dev/random#FreeBSD > 2. Crypto > - Certain protocols have been deliberately weakened, have options that turn > on crypto and/or known to contain backdoors. > > - a robust discussion on how to enable the "highest" standard of encryption > and privacy protective options would be most welcome That is still something that is up for debate. I'm not sure anyone has really accurately identified which are good and which might be compromised from a cryptographic standpoint with high confidence. There are some standards that have been called into question simply because the NSA/DOD/etc recommend them. Are they recommending them because they are strong, or because they have been compromised and they want people to use them? http://www.nsa.gov/business/programs/elliptic_curve.shtml http://en.wikipedia.org/wiki/NSA_Suite_B http://en.wikipedia.org/wiki/Elliptic_curve_cryptography#NIST-recommended_elliptic_curves If compromised ciphers could be positively identified, we could actively discourage their use or disable them as needed. The problem with doing that is compatibility and inertia. PPTP has been broken 100%, but people still use it because they don't want to change, management won't let them change, they have a crazy use case for it, or simply because they don't care. We have placed a large red warning on PPTP for the last few versions and people still keep using it, knowing it's not much better than transmitting in the clear. Jim _______________________________________________ List mailing list [email protected] http://lists.pfsense.org/mailman/listinfo/list
