Max Justicz

Also worth noting that there have been apt RCE's capable of being exploited by a MITM. https://www.debian.org/security/2016/dsa-3733 https://justi.cz/security/2019/01/22/apt-rce.html I really don't see how anyone can still defend not using TLS for debian packages.

Until someone thinks that it should follow redirects which probably leads to the same thing that got apt: https://justi.cz/security/2019/01/22/apt-rce.html Not saying that makes it a bad idea, but importing/downloading trusted code over http(s) is not simple even if the protocol sorta is.

> the overhead The computational overhead is negligible. This simply isn't a credible argument against HTTPS. Netflix send petabytes of data over HTTPS. There's no excuse for anyone else. > A medical site is different to a static html site like listed in the post. Different in degree, but not in category. Your browsing habits on ordinary non-sensitive sites can still be used to profile you. > ISP regardless of SSL know which sites I’m visiting They can still look at the destination IP, yes, and they can probably look at your DNS requests, as secure DNS is currently only rarely used. It's still worth doing. Knowing that someone went on Wikipedia tells you almost nothing. Knowing the specific pages they went to, tells you a great deal. > It’s there in the HTTP host header regardless of SSL or not. That's not the case. HTTPS encrypts all headers. > transit security can easily be encrypted without SSL The easy solution is HTTPS. > people are too lazy to encrypt their content within the application. A universal principle in cyber-security: rolling your own crypto scheme is generally a terrible idea. As I said, Steam and Apt are the exceptions; they use plain HTTP for delivery, and implement secure file verification using hashes. Even here, with competent people running a simple delivery scheme, there can be serious security issues [0]. When it comes to web applications, you cannot implement your own secure delivery, for the obvious reason: an attacker can just…

You mean like - https://justi.cz/security/2019/01/22/apt-rce.html? tbf, I'm being a bit hyperbolic. Its totally reasonable to have security at different layers. One good thing about HTTPS though is it is a very well tested layer. Crypto is such an easy thing to screw up, even when doing just basic stuff.

Similar issues: - Hacking 3,000,000 apps at once through CocoaPods - https://justi.cz/security/2021/04/20/cocoapods-rce.html - Remote code execution in Homebrew by compromising the official Cask repository - https://blog.ryotak.me /post/homebrew-security-incident-en/ - Dependency Confusion: How I Hacked Into Apple, Microsoft and Dozens of Other Companies - https://medium.com/@alex.birsan/dependency-confusion-4a5d60fec610

I found this bug! Here's my blog post about it: https://justi.cz/security/2021/04/20/cocoapods-rce.html I started looking because I wanted to find bugs in Signal for iOS, which uses CocoaPods: https://github.com/signalapp/Signal-iOS/blob/master/Podfile

Here's the kicker: while this might sensible to you, some JSON implementations out there will reject duplicate fields (fail a parse completely), and the RFC does not even specify what is the correct behavior (override previous, ignore duplicates, fail entire parse, return non-unique keyed dictionary, something else entirely?). So while to you and I this behavior might be expected (although I'm still not sure that overriding previous fields is more obvious than ignoring repeated fields) - some library implements thought differently, and there isn't even an agreed on standard. Arguing about this isn't purely academic, either - there have been security vulnerabilities resulting from these differences [1]. [1] - https://justi.cz/security/2017/11/14/couchdb-rce-npm.html

apt (dpkg) binaries need to be anyway PGP signed, so HTTPS is just a another layer of security, which is mostly redundnant. Not too long ago, apt in fact predominantly used HTTP. However, then this MITM vulnerability was found that would've been avoided by using HTTPS, and I think they pretty much switched to using HTTPS from then on: https://justi.cz/security/2019/01/22/apt-rce.html

Because http protocols give you lots of extra chances to screw things up. https://justi.cz/security/2019/01/22/apt-rce.html > Unfortunately, the HTTP fetcher process URL-decodes the HTTP Location header and blindly appends it to the 103 Redirect response:

Does it include the fix for apt's recent mirror redirect vulnerability? If not, any update to a fresh system would be trivially susceptible to a MITM attack if redirects aren't disabled. This includes docker images as well: https://justi.cz/security/2019/01/22/apt-rce.html Canonical really should enable https as an additional layer of protection. This isn't the first time a bug in apt's authentication was found and it's unlikely that this bug will be the last.

Max Justicz post about the vulnerability https://justi.cz/security/2019/01/22/apt-rce.html

The attack can inject fake hashes into the process, so it can pretend the file has the correct checksum: https://justi.cz/security/2019/01/22/apt-rce.html