Re:CloudFlare, SSL and unhealthy security absolutism
by tptacek on Tuesday August 30, 2016 @08:40AM
(#12389439)
If it's that easy, why even have "strict" mode? Why not make the branding less misleading, and rename "Full" to "Weak"?
Re:CloudFlare, SSL and unhealthy security absolutism
by ryandrake on Tuesday August 30, 2016 @09:50AM
(#12390058)
Reminds me of the whole USB 2.0 Full Speed vs USB 2.0 High Speed nonsense from the early 2000s. Raise your hand if you can tell me which one is 480 Mbit/s and which one is 12 Mbit/s without looking it up online.
Re:CloudFlare, SSL and unhealthy security absolutism
by delan on Tuesday August 30, 2016 @10:37AM
(#12390539)
Full Speed has existed since USB 1.0 — it’s “full” relative to USB 1.0, where the only other option was Low Speed — but I agree that the name was a poor choice.
Re:CloudFlare, SSL and unhealthy security absolutism
by jgrahamc on Tuesday August 30, 2016 @08:43AM
(#12389459)
I'll bring that to the PM responsible.
Re:CloudFlare, SSL and unhealthy security absolutism
by epoch1970 on Tuesday August 30, 2016 @09:41AM
(#12389975)
What's the likelihood of actually seeing a change?
Re:CloudFlare, SSL and unhealthy security absolutism
by prdonahue on Tuesday August 30, 2016 @10:09AM
(#12390272)
PM here—he's already brought to me to discuss :)
Re:CloudFlare, SSL and unhealthy security absolutism
by dsl on Tuesday August 30, 2016 @12:12PM
(#12391389)
Can you please also include an HTTP response header that tells us the level of verification of the connection to the origin?
If CloudFlare wants to live dangerously with origin connections, fine... but give end users a way to drop the connection if it isn't secure, like our browsers would normally.
Re:CloudFlare, SSL and unhealthy security absolutism
by toast0 on Tuesday August 30, 2016 @12:21PM
(#12391486)
By the time you've gotten this response header, the request has already been sent over the connection that you don't trust, and at least part of the response.
Re:CloudFlare, SSL and unhealthy security absolutism
by dsl on Tuesday August 30, 2016 @12:22PM
(#12391495)
But at least you can cease further communications. Feel free to suggest a better idea.
Re:CloudFlare, SSL and unhealthy security absolutism
by toomuchtodo on Tuesday August 30, 2016 @07:11PM
(#12394709)
I love your suggestion, with the caveat that browsers send a pre-flight interrogation request (similar to OPTIONS with CORS) to determine if the origin connection is secure before sending a legitimate request (containing potentially sensitive data).
Re:CloudFlare, SSL and unhealthy security absolutism
by DenisM on Tuesday August 30, 2016 @10:51AM
(#12390651)
And is tree cert pinning for self-sign veers? If not, why not?
Or you can bring your own certificate from someone like Let's Encrypt.