nginx2root

Not sure how reliable this is, but might be interest:


"The human researcher was the one that found the specific vuln that affects a default nginx setup with HTTP/2 enabled. This discovery was around the same time this same researcher found another critical vulnerability in a different project that we cannot talk about yet due to the same concerns as nginx2root."

(My emphasis). Wonder if related to the recent Apache 2.4.71 release addressing http/2, or something completely different? Time will tell.
 
Maybe you should try the Caddy web server, if it turns out to be suitable. It's a tough time for C code right now: AI never sleeps, and auditing has become cheaper and easier than ever. And something tells me we'll see a lot more of this
 
Maybe you should try the Caddy web server
Why should he? No need to chase these things or you'll do it all day.

No. It's not a tough time for C code. A vulnerability doesn't care what language it's written in. And exploits are being found quicker using AI apparently.

Why are some treating http2 as an off/on thing? It's a requirement of any today site. It might blow your mind that we were running http3 on our remaining sites a few years ago.
 
The most impactful of the five findings, CVE-2026-123007, allows an unauthenticated attacker to execute commands on an affected nginx server with HTTP/2 enabled. The commands run with the privileges of the nginx worker process that handles the request.

We chained this vulnerability with a separate flaw in the Linux kernel's SCTP subsystem to gain root privileges on the host. In our tests, the chain also escaped a Docker container running nginx and obtained root on the underlying host.

So the whole "2root" part is entirely a chain of linux- and docker-specific flaws/insecurity-by-design mechanisms.... But yeah, nowadays every CVE needs a completely overhyped name. (remember php2whatever? there's not a single working exploit for *production builds* available until today, because the chatbot that "found" the bug always assumed *modified* code *and* an insecure/stupid server configuration for a theroretical exploitation...)

I'm not saying you shouldn't patch, but I'm pissed at all those new chatbot-researchers/slop-kiddies that always overhype their findings instead of keeping it factual and professional. Instead they turn this whole field into a clown-college...
 
Why should he? No need to chase these things or you'll do it all day.

No. It's not a tough time for C code. A vulnerability doesn't care what language it's written in. And exploits are being found quicker using AI apparently.
I said "if it turns out to be suitable," and it depends on the scale: a whole day or a couple of hours. As for the AI, I said basically the same thing, but the misunderstanding was most likely caused by a language barrier
 
quardinimus Again, this is not a FreeBSD issue and not something FreeBSD or anyone else can fix. Once nginx fixes their problem, it shouldn't be more than inserting the code into ports, which will be quick, and recompiling the new code in for packages. Nginx is major software and this won't take long.

I see you are new to FreeBSD and how it works.
 
https requires http2. Browsers will give security warnings if you aren't serving https.
I have HTTPS still with 1.1 and A+ Qualys

alpn.png


Screenshot_2026-11-19_15-10-43.png
 
Back
Top