Re: "Cert spam", or certs with huge numbers of hosts.

2014-10-28 Thread John Nagle
On 10/23/2014 02:00 PM, Richard Barnes wrote: illa and the CA/Browser Forum. And I suspect it is related to this: http://blog.cloudflare.com/introducing-universal-ssl/ I previously wrote "You're probably right". He was. As of the January 2014 U. Mich scan of IPv4 space: Number of IPv4 sit

Re: "Cert spam", or certs with huge numbers of hosts.

2014-10-24 Thread fhw843
o) the whois lists a privacy service in Panama, so who exactly are they?!?   Original Message   From: John Nagle Sent: Friday, October 24, 2014 12:29 PM To: dev-security-policy@lists.mozilla.org Reply To: na...@sitetruth.com Subject: Re: "Cert spam", or certs with huge numbers of

Re: "Cert spam", or certs with huge numbers of hosts.

2014-10-24 Thread John Nagle
On 10/24/2014 06:14 AM, Hubert Kario wrote: On Thursday 23 October 2014 14:30:59 John Nagle wrote: To use Cloudflare you need to transfer the domain to Cloudflare. So it's hardly a MITM. It's a forward proxy service. Not quite. You have to aim the DNS at Cloudflare, not transfer the owner

Re: "Cert spam", or certs with huge numbers of hosts.

2014-10-24 Thread Hubert Kario
On Thursday 23 October 2014 14:30:59 John Nagle wrote: > On 10/23/2014 02:00 PM, Richard Barnes wrote: > illa and the CA/Browser Forum. > > > And I suspect it is related to this: > > http://blog.cloudflare.com/introducing-universal-ssl/ > > You're probably right. What Cloudflare provides by

Re: "Cert spam", or certs with huge numbers of hosts.

2014-10-24 Thread Peter Gutmann
John Nagle writes: >There's a real risk here. A break-in at any of those sites allows >impersonating all of them. This creates a huge attack surface. It's actually a lot worse than that, see "Virtual Host Confusion: Weaknesses and Exploits" by Antoine Delignat-Lavaud and Karthikeyan Bhargavan

Re: "Cert spam", or certs with huge numbers of hosts.

2014-10-23 Thread Matt Palmer
On Thu, Oct 23, 2014 at 02:30:59PM -0700, John Nagle wrote: > On 10/23/2014 02:00 PM, Richard Barnes wrote: >You're probably right. What Cloudflare provides by default is > "Flexible SSL", in which Cloudflare acts as a MITM: Cloudflare acts as a MITM for *all* SSL modes -- because it needs to

Re: "Cert spam", or certs with huge numbers of hosts.

2014-10-23 Thread John Nagle
On 10/23/2014 02:00 PM, Richard Barnes wrote: illa and the CA/Browser Forum. And I suspect it is related to this: http://blog.cloudflare.com/introducing-universal-ssl/ You're probably right. What Cloudflare provides by default is "Flexible SSL", in which Cloudflare acts as a MITM: "For a

Re: "Cert spam", or certs with huge numbers of hosts.

2014-10-23 Thread Richard Barnes
> On Oct 23, 2014, at 4:51 PM, Ryan Sleevi > wrote: > > On Thu, October 23, 2014 1:08 pm, John Nagle wrote: >> Examine the cert of "https://www.sevendays.co";. >> >> Here's one of those certs with a huge number of unrelated hosts. >> This seems to be a Cloudflare legacy setup from the pre-TLS

Re: "Cert spam", or certs with huge numbers of hosts.

2014-10-23 Thread Ryan Sleevi
On Thu, October 23, 2014 1:08 pm, John Nagle wrote: > Examine the cert of "https://www.sevendays.co";. > > Here's one of those certs with a huge number of unrelated hosts. > This seems to be a Cloudflare legacy setup from the pre-TLS era. > Unfortunately, this cert became valid on 10/09/2014. I

Re: "Cert spam", or certs with huge numbers of hosts.

2014-10-23 Thread Matt Palmer
On Thu, Oct 23, 2014 at 01:08:25PM -0700, John Nagle wrote: > Examine the cert of "https://www.sevendays.co";. > > Here's one of those certs with a huge number of unrelated hosts. > This seems to be a Cloudflare legacy setup from the pre-TLS era. > Unfortunately, this cert became valid on 10/09/20

"Cert spam", or certs with huge numbers of hosts.

2014-10-23 Thread John Nagle
Examine the cert of "https://www.sevendays.co";. Here's one of those certs with a huge number of unrelated hosts. This seems to be a Cloudflare legacy setup from the pre-TLS era. Unfortunately, this cert became valid on 10/09/2014. It's not a legacy cert. Should certs like this be rejected as mi