Stories
Slash Boxes
Comments

SoylentNews is people

posted by martyb on Wednesday July 25 2018, @06:07AM   Printer-friendly
from the wasn't-worth-the-work...-until-now? dept.

Submitted via IRC for AndyTheAbsurd

As of today, Google begins shipping Chrome 68 which flags all sites served over the HTTP scheme as being "not secure". This is because the connection is, well, not secure so it seems like a fairly reasonable thing to say! We've known this has been coming for a long time now both through observing the changes in the industry and Google specifically saying "this is coming". Yet somehow, we've arrived at today with a sizable chunk of the web still serving traffic insecurely:

The majority of the Internet’s top 1M most popular sites will show up as “Not Secure” in @GoogleChrome starting July 24th. Make sure your site redirects to #HTTPS, so you don’t have the same problem. @Cloudflare makes it easy! #SecureOnChrome https://t.co/G2a0gi2aM8 pic.twitter.com/r2HWkfRofW

— Cloudflare (@Cloudflare) July 23, 2018

Who are these people?! After all the advanced warnings combined with all we know to be bad about serving even static sites over HTTP, what sort of sites are left that are neglecting such a fundamental security and privacy basic? I wanted to find out which is why today, in conjunction with Scott Helme, we're launching Why No HTTPS? You can find it over at WhyNoHTTPS.com (served over HTTPS, of course), and it's a who's who of the world's biggest websites not redirecting insecure traffic to the secure scheme:

The article continues with a list of "The World's Most Popular Websites Loaded Insecurely", tools and techniques used to gather the data, different responses based on the version of curl, differences accessing the bare domain name versus with the "www." prefix, and asks for any corrections. One can also access the aforementioned website set up specifically for tracking these results: https://whynohttps.com/.

Source: https://www.troyhunt.com/why-no-https-heres-the-worlds-largest-websites-not-redirecting-insecure-requests/


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 0) by Anonymous Coward on Wednesday July 25 2018, @09:50AM (6 children)

    by Anonymous Coward on Wednesday July 25 2018, @09:50AM (#712238)

    But it still would allow users to serve simple static HTML pages without the encryption overhead.

    Enlighten me, what is that bad about encrypting content so that you want us to throw out of the window the benefits of encrypted traffic?
    Are you working for a federal agency too [soylentnews.org]?

  • (Score: 4, Interesting) by Pino P on Wednesday July 25 2018, @01:24PM (5 children)

    by Pino P (4721) on Wednesday July 25 2018, @01:24PM (#712310) Journal

    what is that bad about encrypting content so that you want us to throw out of the window the benefits of encrypted traffic?

    For sites on the public Internet, what's so bad about HTTPS is that there exists no signing-only cipher suite that allows intermediate caching while precluding tampering. If you're serving the same document to a plurality of users, such as serving an encyclopedia article to a classroom full of devices in a school in sub-Saharan Africa with a harshly metered 128 kbps connection, you want a replay attack to be possible. Otherwise, what's the benefit of the HTTP header Cache-Control: public in an HTTPS environment?

    • Cleartext HTTP: Polipo proxy retrieves document once on viewers' behalf and serves it to all 25 viewers.
    • Hypothetical signing-only protocol: Same is possible.
    • HTTPS: Proxy can only process the CONNECT method to make a tunnel through which the same document is retrieved 25 times, once by each viewer's device.

    Sites on a private home network have a different problem with HTTPS. in order to qualify for a certificate, you need a domain name. Let's Encrypt will not issue a certificate if any of the following are true:

    • The certificate is for a private IP address in RFC 1918 space (10/8, 172.16/12, or 192.168/16) or its IPv6 counterpart
    • The certificate is for a hostname within a private top-level domain such as .local or .internal
    • The hostname is within a domain that is not on the Public Suffix List, and 20 other users who have subdomains under the same domain have obtained a certificate in the past week
    • The domain does not support TXT records for the dns-01 challenge, and the host does not accept incoming connections from the public Internet for the http-01 challenge

    So if you can't find a dynamic DNS provider that both is on the PSL and supports TXT records, you end up having to buy a domain name and continuing to pay for its renewal.

    • (Score: 0) by Anonymous Coward on Wednesday July 25 2018, @08:08PM (1 child)

      by Anonymous Coward on Wednesday July 25 2018, @08:08PM (#712630)

      Non-encrypted signing only thing allows for surveillance and data mining.

      • (Score: 2) by Pino P on Thursday July 26 2018, @04:41PM

        by Pino P (4721) on Thursday July 26 2018, @04:41PM (#713210) Journal

        In parts of the world where all Internet connections are very slow and very harshly capped, people are likely to consider "surveillance and data mining" an acceptable tradeoff.

    • (Score: 2) by urza9814 on Thursday July 26 2018, @05:00PM (2 children)

      by urza9814 (3954) on Thursday July 26 2018, @05:00PM (#713225) Journal

      If you're serving the same document to a plurality of users, such as serving an encyclopedia article to a classroom full of devices in a school in sub-Saharan Africa with a harshly metered 128 kbps connection, you want a replay attack to be possible.

      Take one of those laptops and turn it into a caching proxy that drops the encryption. For bonus points, re-encrypt using a self-signed cert that you've already installed as trusted on the remaining laptops.

      Sites on a private home network have a different problem with HTTPS. in order to qualify for a certificate, you need a domain name. Let's Encrypt will not issue a certificate if any of the following are true:

      Why do you need a cert that's trusted on the global Internet for your private home network? Use self-signed certs and install them manually on whatever devices need it. That's a hell of a lot easier than getting a cert from Let's Encrypt or any other CA anyway. I *think* you could also use Let's Encrypt on a free domain like .tk if you configure the redirects properly, but I'm not 100% certain on that.

      • (Score: 3, Interesting) by Pino P on Thursday July 26 2018, @06:13PM (1 child)

        by Pino P (4721) on Thursday July 26 2018, @06:13PM (#713273) Journal

        Take one of those laptops and turn it into a caching proxy that drops the encryption. For bonus points, re-encrypt using a self-signed cert that you've already installed as trusted on the remaining laptops.

        The installation I'm referring to is currently using Polipo software, and Polipo's manual states that it tunnels all HTTPS connections using the CONNECT method. This means we'll have to use something other than Polipo. Which caching proxy software stack do you recommend for terminating HTTPS by issuing a temporary certificate from a private CA and using that to re-encrypt the cached resource?

        Why do you need a cert that's trusted on the global Internet for your private home network?

        Because operating systems for non-PC devices make it painful to install and trust a private CA certificate. A user-installed certificate on Android, for example, won't work in applications designed for Android 7 or later unless the app's developer opts in to trusting user CAs (search keywords: Network Security Config), and it may require changing the lock screen. Some set-top box operating systems offer no way to trust a private CA certificate at all.

        • (Score: 3, Informative) by urza9814 on Thursday July 26 2018, @07:25PM

          by urza9814 (3954) on Thursday July 26 2018, @07:25PM (#713325) Journal

          The installation I'm referring to is currently using Polipo software, and Polipo's manual states that it tunnels all HTTPS connections using the CONNECT method. This means we'll have to use something other than Polipo. Which caching proxy software stack do you recommend for terminating HTTPS by issuing a temporary certificate from a private CA and using that to re-encrypt the cached resource?

          Squidguard can proxy and filter HTTPS traffic so that would probably work...pretty sure you can configure caching on that too although I'm not 100% sure on that point. Looks like Privoxy with Stunnel would also work although that seems a bit more difficult to configure...