- cross-posted to:
- news@lemmy.linuxuserspace.show
- cross-posted to:
- news@lemmy.linuxuserspace.show
You must log in or register to comment.
Outfits that haven’t installed patches since February are getting popped in May by a vuln that was published in January.
Normal technology situations created by normal human behavior. 😜
Outfits? What does it mean in this context?
Organisations
Ahh, thank you
Suits and shit.
Guns for hire
Companies and organizations.
It’s a privilege escalation.
The vulnerability, tracked as CVE-2024-1086 and carrying a severity rating of 7.8 out of a possible 10, allows people who have already gained a foothold inside an affected system to escalate their system privileges. It’s the result of a use-after-free error, a class of vulnerability that occurs in software written in the C and C++ languages when a process continues to access a memory location after it has been freed or deallocated. Use-after-free vulnerabilities can result in remote code or privilege escalation.
a use-after-free error, a class of vulnerability that occurs in software written in the C and C++ languages when a process continues to access a memory location after it has been freed or deallocated.
Immediately I noticed how when Teslas can’t drive themselves we also blame the car and not the driver.
Weak. Blame the driver.
I compiled my own drivers
This guy drives not
And then compacted them!
Yeah, but did you include any biobs?
- LeFantome ( @LeFantome@programming.dev ) 15•10 months ago
I re-wrote my Tesla firmware in Rust. It is faster and more secure. Self-driving is no problem when you use a safe language.
Honestly, why are we even selling cars to people who do not take these basic steps?
Don’t think C / C++ wasn’t blamed.
This is why least privilege is so important. If one account is compromised it will be harder to compromise others if the original account is isolated.
Yet another security issue that Rust would solve.
Oh, we heard, Rust is the greatest invention since sliced bread. We heard it already. Like 65534 times.
Like 65534 times.
So close to full 16-bit max. So close…
Yeah I figured he was going purposely for a memory overflow
Yeah we only need 2 brainRusts more to start seeing some fun.
Gah. I should have stated “I see what you did there.” instead. ;)
Aviation, Health, Space and Car industry have only 3 certified languages that they use. Ada, C and C++. Ada is dying because there are way less young engineers who want to invest their future learning it. Then there is C and C++ but they dont offer memory safety and its really hard to master and its really hard and long (thats what she said) to certify the code when being audited for safety by a tier company.
Rust solves by default (no need to review) like 2/3 of the standard requirements those industries have and are that found in C and C++. Rust will soon be approved in this group by the car industry.
Im not a rust fan, but I have 3 things to say about rust.
- Its fun to program like C++ having the peace of mind knowing the compiler is there helping.
- You dont feel like youre defusing a bomb like when writing C.
- Even though its a fun language to write, its also really hard to master, itd say 2 years to be really proficient with it. There is just so much knowledge.
Aviation, Health, Space and Car industry have only 3 certified languages that they use. Ada, C and C++.
Rust is automotive certified since over half a year. https://ferrous-systems.com/blog/officially-qualified-ferrocene
- You dont feel like youre defusing a bomb like when writing C.
Whoa, Skippy. It’s not saving the world, it’s just coding properly.
Well no, those companies deal with really important subjects. Airplanes, car safety, chemotherapy machines, missiles, etc. Have a good day
Could you explain the “no need to review” part? I do keep hearing good things about Rust.
These industries hire third parties to review c and c++ line per line to make sure it’s memory safe. Rust by default forces you to write memory safe code, otherwise it won’t even compile. The rust compiler tells where is the problem and what it expects. No only for basic Type errors but also for concurrent code.
Is it not possible to build that functionality into C/++ compilers?
its the way the language was built. Im not sure its possible without breaking C/C++ which have like 35 years + in the making. Also these concepts are have little to do with programing and more architectural designs. The designers are real engineers working on difficult concepts. All big brains tbh
Ada SPARK is not dying at all, it’s growing. It is used where formal proof is required like and Rust is nowhere near that!
I wonder how many folks are just refusing to use Rust to spite the Rust Evangelism Strike Team.
Rustaceans 🤝 Vegans
I wish there was a synonym for “evangelism” that began with a “u”.
Urge? Kinda dark and villainous feeling.
Upgrade! “The Rust Upgrade Strike Team! Upgrade Today!” Sounds very propagandistic, almost doublespeak.
Ultimatum? Mildly threatening.
Utopia? It has the self righteous feel.
Uhvangelism, hurhur.
Universalism?
I giggled, thank you.
user
I wait until cargo is actually secure.
What is insecure about it?
It doesn’t verify downloads are authentic. Its an issue with almost all programming dependency managers besides mature ones like Java’s Maven.
Python has been working with Facebook to fix this in pip for like a decade.
But obviously it shows that rust isn’t so concerned about security.
Ah interesting. Thank you, you’re giving me something to read about that I never considered for crates. I guess I just assumed because of the scrutiny Rust was built with and continues to go through that it would also apply to verifying crates. I have definitely heard about it with NPM so it should have been obvious that it might not be any different for crates. Thanks again!
I hate it when people talk about new technologies 🤬
Same. We should head back to ICQ!
eh, still beats Discord as far as I’m concerned
Yet another problem that actually updating your shit - which is trivially easy on enterprise Linux - would fix.
It’s part of the 95% of problems solved by actually updating your enterprise Linux host.
unattended-upgrades and forget about it
Tell me more (for real, I’m unfamiliar).
Its a Debian package that automatically upgraded packages (if they have pending security updates)
I run mine manually, good to know. Will check it out.
That requires that the patches be in the repos. With RHEL it might be a few months
Any software can have security issues, including ones written in rust. Just because C/C++ allows one to shoot oneself in the foot doesn’t mean it’s something that’s commonly allowed by anyone with any skill, it’s just a bug like anything else. I swear, people advocating rust believe that it’s something intrinsic in C/C++ that allows such a thing regardless of what a developer does, and it’s getting tiresome.
Of course a good developer can avoid these problems for the most part. The point is that we want the bad developers to be forced to do things a safe way by default.
But it is, do you not understand what rust brings compared to these two languages ?
A language for noobs that encourages bad style and programming because you can’t shoot yourself in the foot as easily (but you totally still can)? That’s what all these fad languages seem to be, and more keep popping up and declaring themselves the future of programming all the time. Just wait, rust will be forgotten for some other fad language everyone will start using soon enough. Stop reworking everything into the fad language of the moment and just work on existing code.
I’m sorry but this reads like someone that hasn’t used Rust or hasn’t spent much time with it. You’re generalizing Rust with other languages while forgetting that some fads turn into standards.
If everyone stopped trying new things we’d never see progress.
Edit: fixed typo
A language for noobs
That assertion surprises me; I find C easier to use than Rust.
The problem is bad programmers. You can write good C code but it takes more effort and security checking. You also can write vulnerable and sloppy Rust code.
I don’t think it’s realistic to expect a rewrite of code that works. Maybe over time we can start implementing pieces in safer languages.
I admit C++ ain’t safe, but wonder if there’s an alternative to going Rust. Don’t get me wrong, I love the language. But Rust is a beast on its own. I read here that game devs generally can’t adapt Rust because the language forces frequent refactoring, which doesn’t fit the business speed of game development.
I don’t care for Rust because I like writing unsafe code. It’s fun. However, I would value the assurances it provides using software written in Rust.
Is there a way to jailbreak an Android phone using this exploit?
You could just unlock the bootloader
Assuming the bootloader is unlockable
You could just buy an android phone that encourages this. All Pixels, for example.
+1 for Pixels but -1 for Google’s “support”. You’ll never talk to a human with them. I love GrapheneOS on my Pixel though, and they’re really the only phones you can install it on cause you can re-lock the bootloader on it after installing. CalyxOS (fork of Graphene with slightly less sandboxing) does support FairPhone 4&5 and a few Motorola phones though.
Is this even new?
I thought this already circulated a few months back.
Even Debian stable has already patched it.
Debian is actually one of the fastest patchers
RHEL on the other hand
Security patches do the opposite of break stuff
This is the best summary I could come up with:
It’s the result of a use-after-free error, a class of vulnerability that occurs in software written in the C and C++ languages when a process continues to access a memory location after it has been freed or deallocated.
At the time this Ars post went live, there were no known details about the active exploitation.
A deep-dive write-up of the vulnerability reveals that these exploits provide “a very powerful double-free primitive when the correct code paths are hit.” Double-free vulnerabilities are a subclass of use-after-free errors that occur when the free() function for freeing memory is called more than once for the same location.
The write-up lists multiple ways to exploit the vulnerability, along with code for doing so.
The double-free error is the result of a failure to achieve input sanitization in netfilter verdicts when nf_tables and unprivileged user namespaces are enabled.
Some of the most effective exploitation techniques allow for arbitrary code execution in the kernel and can be fashioned to drop a universal root shell.
The original article contains 351 words, the summary contains 168 words. Saved 52%. I’m a bot and I’m open source!