News for hackers. Stuff that matters.
Re:Synology Security Advisory by RetroTechie on Thursday November 19, 2026 @07:17PM (#50568467)
I've always wondered: if you take the cause of each CVE, look at underlying cause of that & repeat, how many essential root problems would you be left with?

"Can't do math" (off-by-1 errors etc)

"Sanitize your inputs" (escape sequences, parsing etc)

"What are you referencing?" (null pointers, use-after-free etc)

and error handling. Plain error codes bubbling up a call stack, or bypassing that somehow which may violate program structure.

How many of these fundamental issues are there?

Re:Synology Security Advisory by Waterluvian on Thursday November 19, 2026 @07:20PM (#50568507)
“In the beginning the Universe was created. This has made a lot of people very angry and been widely regarded as a bad move.”
Re:Synology Security Advisory by chias on Thursday November 19, 2026 @08:04PM (#50568848)
Boil it down enough and you get four:

1. Failure to differentiate data from instructions

2. Business logic flaws

3. Insecure operating environment

4. Incorrect use of cryptography

Re:Synology Security Advisory by kurthr on Thursday November 19, 2026 @08:52PM (#50569172)
LLMs hits all of the top 3!

I mean, I'm kinda kidding, but #1 is #1.

Re:Synology Security Advisory by intrasight on Thursday November 19, 2026 @09:46PM (#50569533)
Von Neumann architectures means number 1 is unavoidable
Re:Synology Security Advisory by silon42 on Friday November 20, 2026 @03:41AM (#50571862)
1 can be mitigated by appropriate use of page protection flags (write or execute).

Execute without read is not supported so much though.

Aside from language level enforcement (Java/Rust/...)

Re:Synology Security Advisory by TeMPOraL on Friday November 20, 2026 @03:49AM (#50571937)
Reality means number 1 is unavoidable in general, for fundamental reasons. Our physical universe has no "control"/"data" distinction.

It's what still tripping half the people working on security when it comes to LLM.

Re:Synology Security Advisory by meindnoch on Friday November 20, 2026 @05:11AM (#50572582)
Data should be using binary, but instructions should use ternary. Then they wouldn't mix.
Re:Synology Security Advisory by agos on Friday November 20, 2026 @05:51AM (#50572903)
I'm fairly sure Turing and Gödel means number 1 is avoidable
Re:Synology Security Advisory by guax on Thursday November 19, 2026 @10:39PM (#50569851)
You can simplify it even more!

1. Error that causes security problems

Re:Synology Security Advisory by TeMPOraL on Friday November 20, 2026 @03:53AM (#50571966)
And then you're maybe finally forced to answer the two most important questions that security-minded people love to pretend don't exist:

What is being secured (as in, whose interests), and from who?

Security is all nice until one day you find yourself being seen as threat actor. See e.g. all the modern computing devices you "own".

Re:Synology Security Advisory by kalaksi on Friday November 20, 2026 @08:28AM (#50574369)
Not sure what you mean. Threat modeling is pretty well known practice.
Re:Synology Security Advisory by rpdillon on Friday November 20, 2026 @11:09AM (#50576919)
He's alluding to the fact that device manufacturers want to secure the device from the device owner.
Re:Synology Security Advisory by solyin on Friday November 20, 2026 @11:54AM (#50577799)
Apple's done a pretty good job of it!
Re:Synology Security Advisory by tensegrist on Thursday November 19, 2026 @08:45PM (#50569123)
Re:Synology Security Advisory by red-iron-pine on Friday November 20, 2026 @12:48PM (#50578878)
came here to say this ^.

OWASP already covers this

Re:Synology Security Advisory by naasking on Thursday November 19, 2026 @09:12PM (#50569320)
> I've always wondered: if you take the cause of each CVE, look at underlying cause of that & repeat, how many essential root problems would you be left with?

It's almost always memory safety violations. The rest are basically application-level bugs, like not properly checking inputs.

Re:Synology Security Advisory by bluGill on Thursday November 19, 2026 @10:03PM (#50569646)
It is almost always social attacks.

CVEs are a big deal, but humans are the real vulnerability

Re:Synology Security Advisory by lofaszvanitt on Thursday November 19, 2026 @11:39PM (#50570246)
grsec
Re:Synology Security Advisory by catlifeonmars on Friday November 20, 2026 @12:04AM (#50570387)
[delayed]
Re:Synology Security Advisory by flowerthoughts on Friday November 20, 2026 @01:28AM (#50570886)
> What are you referencing?

I'd phrase that as "know your data access ordering," and include races.

Re:Synology Security Advisory by fulafel on Friday November 20, 2026 @01:36AM (#50570942)
These are all bug classes that are handled by reasonable programming languages since decades ago, so can be reduced to not taking risk management seriously.
Re:Synology Security Advisory by rtpg on Friday November 20, 2026 @02:13AM (#50571186)
time of check/time of use issues _might_ be considered "What are you referencing" but the arrow of time feels so important there that I'd consider it its own class
Re:Synology Security Advisory by wccrawford on Friday November 20, 2026 @09:15AM (#50574944)
"sanitize your inputs" means you have to know what the output is ahead of time, and that you'll never output any other way. It also means false positives. It can actually be harmful if done wrong (double escaping, escaping that certain systems don't understand) but there's some basic forms that probably aren't too bad. It's generally pointless, though, since...

"escape your outputs" works for all data and is actually achievable. It means even escaping data that was never an "input" from the outside the system, even developer-created data, which absolutely can (and does) cause problems to ignore.