agwa

62 ms·

721.
by agwa 11y ago
Alas, random serial numbers are only a "SHOULD" in the Baseline Requirements, and not all CAs use them.
722.
by agwa 11y ago
> Facebook's proposal is that only organizations that are a) demonstrably using this technique and b) validated at a higher level than automatic domain validation be allowed to get new SHA-1 certs b) may help prevent collision attac
723.
by agwa 11y ago
> Which means adversaries could break the encryption, but could only afford to attack a small number of people No, because an attacker could forge a single CA certificate that could then be used to sign certificates for an unlimited numb
724.
by agwa 11y ago
> In this particular case, Facebook is advocating for using a SHA-1 certificate only if they know that the client doesn't support SHA-256, and if that knowledge is cryptographically protected. Since TLS supports a way of securely de
725.
by agwa 11y ago
> (it's undefined for unsigned integers in C): No, unsigned integer overflow is defined to wrap around in C. Only signed integer overflow is undefined. > I wonder why Go doesn't just define signed integers to be in 2's
726.
by agwa 11y ago
Unfortunately not. STARTTLS is the bane of standalone TLS terminators like titus, which is one of the reasons I really dislike STARTTLS. I won't rule out titus supporting STARTTLS some day, but the idea of integrating parsers for a b
727.
by agwa 11y ago
There's a cost to the isolation. It's not for everyone. Ultimately, I'd like to provide a knob to the administrator to control how much isolation they'd like. If you can't afford full isolation, you could tell tit
728.
by agwa 11y ago
(Author here.) I'd be more wary of using security software that changes frequently, since every code change is an opportunity for a new security vulnerability to be introduced. I'm very cautious with changes to titus. That said,
729.
by agwa 11y ago
The name is inspired by stud (Scalable TLS Unwrapping Daemon), another TLS termination server with a cool acronym.
730.
by agwa 11y ago
Except that Microsoft hired people out of college to work on Midori, which they surely wouldn't have done if Midori was simply a place to dump the people they couldn't/wouldn't fire, which was the point of DEC's &qu
731.
by agwa 11y ago
I believe the point tedunangst is trying to make is that if the API works something like this: size_t sz = lib_get_buffer_size(); char *buf = malloc(sz); lib_fill_buffer(buf); and there's a bug in lib_get_buffer_size that ca
732.
by agwa 11y ago
I'm all for rolling out opportunistic STARTTLS, but I don't see how you can surface it in the UI in a way that's meaningful to the vast majority of users. If STARTTLS was not used, the message might have been intercepted. If
733.
by agwa 11y ago
If Gmail did this, the MitM would be sure to correctly validate Gmail's certs.
734.
by agwa 11y ago
> At least it wouldn't be intercepted on that hop It wouldn't be passively intercepted. It could trivially be actively intercepted. Gmail would not display a warning in this case, thus making the user think their password&#
735.
by agwa 11y ago
This is a bad move that will give users a false sense of security. Server-to-server SMTP is unauthenticated. Just because an email arrived at Gmail over an encrypted connection doesn't mean that the connection wasn't intercepted
736.
by agwa 11y ago
3DES (the cipher) is secure but incredibly slow. It's often included in server ciphersuites to support old clients (the alternative for old clients is RC4, which is not secure). Edit: I should mention though that 3DES as used in TLS i
737.
by agwa 11y ago
There's nothing inherently wrong with the DHE ciphersuites, as long as the server provides secure parameters. Since weakdh is really a server-side issue, and howsmyssl.com is a client-side test, a warning doesn't really make sens
738.
by agwa 11y ago
OpenVPN makes you provide your own parameters; there is no default. As of Apache 2.4.7, the default DH parameters have the same number of bits as your RSA key, and since CAs have required at least 2048 bit RSA for a few years now, you'
739.
by agwa 11y ago
The OpenVPN section is misleading. The dh option is only supported on the server side. If you try to use it on the client side (which is what this guide appears to be tailored towards) it will be ignored and you'll use whatever DH pa
740.
by agwa 11y ago
Note that the findings referenced by your first link are total B.S.: http://blog.erratasec.com/2015/04/no-75-are-not-vulnerable-t...
741.
by agwa 11y ago
Indeed. I wasn't implying that the blog post authors were copycats. I've updated my comment to hopefully avoid confusion.
742.
by agwa 11y ago
They published a technical report in May that is almost identical to the paper submitted to the conference. No criticism of the authors is intended (to the contrary, their findings are amazing and very important). But readers need to know
743.
by agwa 11y ago
There aren't any new findings here. It's merely a rehash of the Weak DH attack (by the same researchers) that was made public in May of this year: https://weakdh.org/ Still, it's a good reminder that you sho
744.
by agwa 11y ago
Yeah, I just emerged from the Git code myself ;-) So you're correct that GPG-signing commits (but not tags) prevents collisions in commit objects. The problem though is that a commit ultimately contains a SHA-1 hash of a tree object,
745.
by agwa 11y ago
GPG can use SHA-2, but the way that GPG signing works in Git is that you essentially GPG-sign the SHA-1 hash of a Git object. So even if GPG is using SHA-2 for its signature, it's no protection against a collision in one of the underl
746.
by agwa 11y ago
How does GPG-signing help, when the signature is over a SHA-1 hash?
747.
by agwa 11y ago
> It's not because FLIF is GPL v3+ now, that we can't add more permissive licenses later. That's only guaranteed if you obtain copyright assignment from any contributors now . Otherwise, re-licensing will be a massive he
748.
You Can Now Turn Off Ads on Techdirt(techdirt.com)
4 points by agwa 11y ago|0 comments
749.
by agwa 11y ago
From `openssl s_client -connect signed.bad.horse:443 -servername signed.bad.horse` (scroll the box to the right, and read from bottom to top): Certificate chain 0 s:/C=US/ST=Minnesota/L=Minneapolis/O=Sandwich.Net,
750.
by agwa 11y ago
Automotive security is such a terrible example for the security/usability tradeoff. Cars would be no less usable, but incredibly more secure, if they simply isolated the brakes on a separate CAN bus. Security/usability tradeoffs

More ›