Skip to content
  • 0 points on Sorry, Humble. You lived long enough to see yourself become the villain. in !games@lemmy.world ·

    Just by eye I think I can make it out. The domain looks like [check dms] to me.

  • 2 points on What was the silliest issue you discovered during debugging? in !programmer_humor@programming.dev ·

    Redefining the number 2 in Python

    The biggest const-by-default lang is probably Rust now isn’t it?

  • 5 points on What was the silliest issue you discovered during debugging? in !programmer_humor@programming.dev ·

    Might be the character count. Wednesday has many, September has many, after the 9th has two.
    Program filling the date into a form field of width 1 less than was possible with actual dates.

  • 2 points on What was the silliest issue you discovered during debugging? in !programmer_humor@programming.dev ·

    Hope you never have to use a language that lets you redefine numbers.

  • 0 points on Questioning everything in !comicstrips@lemmy.world ·

    Edit: I’m an idiot, I thought by “found, ¬generated” you meant you found it, and this is not generated (not ai).

    Hows the hair work? Are we actually sure this isn’t ai?
    I tried tracing it and it seems to not go back very far. This pinterest one is the oldest I’ve seen, and it is marked as ai modified.

    I’d say this strongly looks like ai cause it is 2024-5 era ai.

  • 6 points on NGINX2Root by Xint — From nginx visitor to root of the server host in !selfhosted@lemmy.world ·

    Even calling this “nginx2shell” would be very clickbaity. This isn’t a realistically usable attack. Only one exploit can ever be a “2shell” for any software, it should be a once in a decade vulnerability getting that name. Even if it was escalatable, and even if it didn’t require extensive resources for every exploitation, it’d steíll depend on only semi-common configuration. Nothing deserving of “2shell” on multiple levels.

    With the privesc, I would look into that again. I got the impression we were through the major escalation pathways, and now have increasingly convoluted and less plausible leftovers being reported.
    Even if they report 1k garbage exploits at once, that’s still not gonna do much for any actual real-world risk to normal hosters.
    Given the marketing they do when something is even theoretically plausible as an escallation, and given the worst I recall hearing recently was one of the tcp ones that only “worked” for sandbox, container, and vm escape (and ofc user to root), I don’t think we have that many more left. Maybe ai companies are holding a few back for another big press release or two, we’ll see. If this bundled root exploit turns out to be a major dud, I doubt they have much material left.

    AI companies have an interest to embed the general sentiment that ai exploitation is growing exponentially, and the only way forward is to buy ai and keep testing and developing software with it to protect from ai attacks ahead of time.
    But in practice in the existing fields ai was introduced to, there was a saturation. Initial gains when making ai capable of working in the field and it then jumping from a few categories to most inside that field. and then some more for more specific training for the tasks, and then not much past that.

    We had a few classes of bugs AI was suddenly very capable at finding. We probably got all the faulty page handling in spring, then various use after free and general memory handling bugs. We’ll probably get ai implementing existing testing suites more widely across software. Like how the 4-cert was found by just running the normal testing tool on some random cert companies obscure intermediate certs, or how some basic sanity checks found the musl ssh password thing.
    Once memory saftey has been improved to the level current llms can understand, and basic security tests have been written for software, we might extend this to continuously more obscure software but eventually there’d not be much more to report.

    My expectation is AI companies are stockpiling the stuff they have, releasing it on a schedule and with marketing that suggest an ever worsening problem, and then hope enough people buy ai products that they can then blame the sudden reduction on ai found problems on the new ai testing.

    To be fair, this is a bit conspiritorial, but even taking the exact opposite view, there’s still only so many bugs to find. If you think ai will keep growing in bug finding ability, then at some point they’ll find all, and the open weight models will also find all, and then eventually this all becomes a unit test that runs before stuff gets pushed.
    In my own expected scenario where llms are good at basic memory safety and logic bugs, that’s equally be the end result. Using some specialized local model as a compile-time check stand-in for something like the rust borrow checker or lean logic checker for languages that suffer from shit memory protections (looking at C++) or encourage bad logic patterns (looking at basically anything other than proof checkers and maybe rust).

  • 70 points on NGINX2Root by Xint — From nginx visitor to root of the server host in !selfhosted@lemmy.world ·

    Just fyi, the nginx vulnerabilities are all impractical to use. The bugs are use after free, which will almost definitely lead to a crash in the worker process and have them be restarted.

    That itself would make it about as bad as the mp-tcp style expoits lately, but it’s even weaker. If you could keep trying, you could strap an llm into figuring out a specific machine and after tons of compute eventually get out data or take controll of the nginx-owned process.
    But, the restarts should restart that process, so I don’t see almost any risk unless you have address randomization disabled, which would be highly unusual.

    Unless I am missing something here, I see a bigger risk with an unmentioned loop that was also fixed in nginx. That one lets you make a worker thread hang, so you can disable a server until someone notices and restarts nginx. That ddos affects servers with tls1.3 support.

    Also their “2root” is just a completely unrelated kernel bug they bundled for marketing, which is pissing me off tbh. All of this “security research” for marketing is the worst possible contradictory incentive. You wanna release as poorly as possible to maximize fear mongering.
    They didn’t even release the bug, for all we know is as convoluted to use as the mp-tcp stuff and affects basically noone. Not that the nginx side is exploitable.

  • 4 points on Quad9’s withdrawal from France: A canary in the coal mine in !technology@lemmy.world ·

    IP certs are quite hard to get, used to be basically impossible not too long ago.

    Edit: Not so hard anymore. Letsencrypt added them this year (finally) with the normal http challenge and 6 day lifetime. Which should be fine in normal automated setups. Still not supported accross major acme providers but this’ll make it easy enough to do.
    So TLS on bare IPs should be just fine now.

  • 4 points on Punk/Noise rule in !onehundredninetysix@lemmy.blahaj.zone ·

    Just fyi, you linked that with the tracking part included, which shares your youtube account and somehow makes youtube start a DM chat to you when clicked with this video as the first “message”, and thus lets anyone DM you.

  • 1 points on The Invention of The Wheel in !comicstrips@lemmy.world ·

    Mill stones don’t have to be round do they?

  • 4 points on NoisyTCP: Linux Kernel TCP Pre-auth Remote Corruption in !linux@programming.dev ·

    Reducing the RX queues of the interface where external packets come in to one and disabling RPS and RFS

    This should already be the case on normal systems. As I read this, if a system doesn’t have a server grade nic and is using custom values or bonding interfaces, it shouldn’t be affected.

    An actual attack would also realistically need access to the local network to have a chance at the right timing, and would only achieve a crash. Unless you let a high end ai run for hours cracking at that one specific system and network scenario presumably, where, with intranet access, you could then maybe get rce.

    If you feel worried you could set sysctl net.ipv4.tcp_syncookies=0 Which is their other recommended mitigation. Though then you’d be open to syn flooding, which is way easier to do, so I don’t see the point.

  • 1 points on Voyager I will have traveled one light-day on Tuesday. in !mildlyinteresting@lemmy.world ·

    DP’s in usb4. hdmi can step on a rusty spoon and die of scurvy for all I care.

  • 2 points on Kid named Uncle in !tumblr@lemmy.world ·

    Alternatively it impregnates you and cuts you in half.

  • 5 points on Yes, you are! in !onehundredninetysix@lemmy.blahaj.zone ·

    it’s simply not possible! bottoms are pure of heart, and mine is a poisonous stockpile of hurt and malice.

    The most bottom statement in the history of language.

  • 0 points on What are some things X11 can do that Wayland still can't? in !linux@programming.dev ·

    That seems odd that it wouldn’t happen on x11 on the same settings then. I’d place the blame on your window manager.
    Mine at least has a setting for showing icons instead of the window contents, on both x11 and wayland. If I had a program without icon woth that setting I’d get a big icon no matter what.

  • 0 points on What are some things X11 can do that Wayland still can't? in !linux@programming.dev ·

    I’ve only seen them on the taskbar for niche giu programs that don’t set an icon. Where do you get a big wayland placeholder?

  • 0 points on What are some things X11 can do that Wayland still can't? in !linux@programming.dev ·

    Whats your lemmy app? I assume it doesn’t like svg

  • 8 points on Why is all air traffic in the US currently grounded until further notice? in !asklemmy@lemmy.world ·

    Screw the Carter admin for purposefully installing slow routers to sabotage future ai innitiatives.

  • 1 points on What are some things X11 can do that Wayland still can't? in !linux@programming.dev ·

    Xfce does support wayland, though in earlier stages than most, who had it somewhat stable it for like 3-4 years now. Xfce started experimentally with 4.20, and should be stabelizing it right now. If you update to 4.22 unstable it might just work just fine.

  • 1 points on What are some things X11 can do that Wayland still can't? in !linux@programming.dev ·

    That’s ok. It’s ok to be wrong.

    X11 wayland
    X11 replacement icon wayland replacement icon

    I mean they are both really annoying.