Slashdot Mirror
Linux Kernel Back-Door Hack Attempt Discovered
An anonymous reader writes "The BitKeeper to CVS gateway was apparently hacked in an attempt to add a root exploit back door to the Linux kernel, according to the linux-kernel archive. The change was in the file kernel/exit.c and changed the user ID of a process to root under the guise of checking the validity of some flags. The core Linux BitKeeper kernel repository was not at risk, and in fact it was the BitKeeper CVS export scripts that detected the unauthorized modifications to CVS. The changes were falsely attributed in CVS to long-time Linux developer davem (David Miller). Users of the BKCVS repository should resync their trees to remove the offending code if they had replicated it since yesterday."
Good to see the system works. You would wonder what would happen if said hacker was working for a company on a similar closed source program. Would it have been detected?
Someone has some damned big balls to do something like that...
Let's hope they're cut off.
This statement is false.
Anybody point fingers at Microsoft yet? SCO?
This is horrible. Thank heavens BitKeeper recognized the failed attempt to alter the code of the kernel. This is something the kernel developers should really come up and implement a plan to maintain kernel "safety".
YOU'RE WINNER !
Another lame blog
This is the reason I trust open source software. The power of peer review (in one form or another) catches these kinds of things before they are sent into the wild.
Imagine if this had sneaked into some Longhorn code right before shipping. Many eyes make few mistakes.
this sig limit is too small to put anything good h
Sounds like a plan to get the dirty GNU/hippies to upgrade to the real BitKeeper instead of using the communist CVS gateway.
That McVoy is a smart one!
Did you know his programmers need to feed their families and pay their mortgages? Very sad situation, I hope everybody buys 10-15 licenses ASAP.
Why not just establish a web-o-trust and sign patches?
That way people who hack in won't be able to send in signed patches to the system [e.g. even if they physicially update the tree others can trivially spot the unsigned patches].
That would of course, require people to actually think about security in terms of "oh sure people won't hack it because it hasn't been done...much...before."
Tom
Someday, I'll have a real sig.
i want to know if the hack was a remote backdoor or "only" a local root compromise. In order to how bad was the hacker that try to do this.
Thanks to the admins and developers that detect that!
Damia
Hats off to the people who check and report this kind of stuff.
+-+-+-The folowing statement is true. The previous statement is false.-+-+-+
Will this be the first of more kernel backdoors, now that the idea is out there?
That would be giving them.........way too much credit.
I'll call ESR, he's got the guns.
You guys get Linus and make sure he brings Tove, since she could probly kick all our asses.
Once thats done we'll Larry McVoy, by this time hopefully he will have the IP of the slimeball.
The Pose rides at Dawn, we can kill some Trolls along the way.
Did Glenn Beck rape and kill a girl in 1990? gb1990.com
Ok, so the scripts caught an attempt to install a back door. Everybody jumps up and down and sings the praise of the mighty Open Source Movement.
What if a backdoor was installed last week, or last month, but was not caught?
The fact that this was possible once, should really make people think about the possibility of it happened ALREADY, and determine if it is necessary to hunt through the code for a systematic review.
Instead, all we get is Microsoft Bashing...
Ugh
In my code I always put the constant on the lhs so that the difference between the equality (==) and assignment (=) operator are caught by the compiler by accident.
if ((options == (__WCLONE|__WALL)) && (current->uid = 0))
In this case, it would make an attempted root hole more visible, as (0 = current->uid) would not compile.
You honestly have no idea what a "monolithic kernel" is, do you.
Or HIBT.
And this is exactly why folks should insist on open source code.
Assuming it was noticed, and I have little reason to think that modification of a project's cvs tree would go unnoticed, a closed source product would have to go up and down the development chain of command. Then likely up and down the marketing chain of command while a decision was made whether to say anything about it (yeah, right) was made. Meetings would be held, blame would be assigned, and - oh yeah - a discussion about a fix would ensue.
Perhaps I exaggerate, but only a little.
I remember when a beta of a game [unnamed software publisher] was working on got ripped off our company ftp site and passed around. There was so much hype about our game that the leaked late beta was a serious disappointment and effectively killed the good buzz the marketing folks had whipped up. [It blew anyway, got shredded by the gaming rags, had a lot of potential but an inexperienced crew and very little financial support.]
Of course, this situation is nothing like that.
There's always going to be someone trying to backdoor the linux kernel, windows, osx, apps galore. Having the source on-hand to look at gives you that added level of confidence that "hey, worst case we can fix it - deal with it ourselves" rather than go through the denial, silence, lame excuse, patch cycle you go through with closed source products.
In other words: 1) Work on the code for a long time, developing good features and build up virtual reputation points so that people trust you. 2) One day decide to insert your backdoor amidst some big checkin. 3) Disappear.
It doesn't seem hard for someone to pay some random third world programmer to do this so. For example, if Red Hat had a guy in russia doing this they could, after the latest kernel was widely distributed, use it to attack Novell/SUSE.
Think about how often your own code gets reviewed, debugged, optimised - not necessarily by you, but by Joe Coder on the same project or Wilma Coder some time down the road.
I doubt that someone sociopathic enough to work on something for years under a legitimate guise, then "one day" would be able to keep it together long enough to pull off the kind of coup you're proposing.
No, I tend to think that folks of this bent are found mostly among the crowd of virus authors.
When you troll like that, I think you're supposed to have some throw-away account so you can collect karma in some misguided intent to abuse the moderation system. I hear that's what all the kids are doing these days.
:)
(wait - am I supposed to say "here goes my karma" at this point?)
You can do the same thing in a few other operating systems. I believe one of them is a unix of sorts but I do not know which one nor have I gone to look for it.
My God! It's full of stars!
1 x 4 x 9
That monolith... oh... kernel.... right...
Stop the Slashdot effect! Don't read the articles!
Keep reading the thread - you'll read a bid of Linus, and a comforting explaination of whats happenin' back there.
The Ultimate Backdoor, if anyone hasn't read about it:
Reflections on Trusting Trust.
You might want to doublecheck that gcc code you're compiling the kernel with...
Those of you not as well versed in the lore as the rest of you would do well to read the legend of Ken Thompson's backdoor login compiler compiler.
Waiting for ad.doubleclick.net...
In my code I always put the constant on the lhs so that the difference between the equality (==) and assignment (=) operator are caught by the compiler by accident.
Good style.
But this was apparently not an accident, but a deliberate attempt to disguise a trapdoor. As such the author would, of course, just "forget" to use that piece of defensive programming. B-)
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
The "backdoor" that someone attempted to submit was a local privilege elevation bug, not a remote compromise.
Your childish tricks will not work around here.
And please return the check to Bill, you failed.
Will this be the first of more kernel backdoors, now that the idea is out there?
Isn't the pertinent question... was this the first?
Sounds like it might be time for a quick audit. Like looking at every instance of "current->uid" to see if the uid is a LHS.
Don't forget things like "current->uid -= current->uid" and similarly with exor, bitwise and, etc. (That's why any occurrence of current->uid as a LHS is suspect.)
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
The problem is that CVS was exploited.
Not true, BitMover infrastructure has been compromised. We don't know how this was done. If it was done through CVS, it's BitMover's fault anyway (CVS is notoriously insecure, especially in pserver mode).
what a piece of crap you wrote man..
Absolutely. Where did these lines come from? What computer was compromised? Was cvs pserver the security hole?
--jeff++
ipv6 is my vpn
No more Linux for us
Yeah, because he'd rather like a closed source product where such attempts suceed unnoticed.
Do you care about the security of your wireless mouse?
Isn't the pertinent question... was this the first?
No. I'm pretty certain that the CCC once smuggled in a backdoor to the login prompt. That must have been pre 1994.
I can't find any more information on this. Does anyone have a link?
Insightful my ass. Nowhere does it say CVS was exploited.
The code was injected into a CVS tree, the box could have been compromised in another fashion, such as a wu-ftp hole or some such thing.
So please, don't throw the word exploit around as if you have 1/2 a clue about security. It just makes you look silly to those of us who do.
If you change that to if((c=1));, the error will be supressed - the extra parenthesis show that the assignment was intentional. Which it was, in a way.
Litigious bastards
We'll see if they pounce on this. If so, expect something like "See? Open Source projects are inherently vulnerable to hackers infiltrating and corrupting the code base" This is not good. But it is a good thing it was caught and never got released!
--
om Shanti
I see a lesson in this: The oldest tricks are probably the best, or else they wouldn't live long enough to be "old".
I'm a LKML subscriber, so yeah I'll say this: Larry McVoy's gonna be pissed, and $DEITY help whoever...
If you think the flamewars and trolling is nasty here, you should see them over there, trust me on that.
C|N>K
Thank heavens for a BitKeeper. I'm glad they took measures to secure such a thing from happening before it did. Only smart minds in the linux community could do something like that. Screw closed source.
I always sort of wondered what was stopping people from putting a "virus" or something into the portage tree that would knock out my system with I updated it.
Unfortunatly I don't think it would be very hard for someone to insert something that could totally hose my system.
This kinda scares me.
-Mary
I copied from here...
--------
Bleah! Heh heh heh... BLEAH BLEAH!!! Ha ha ha ha...
how do you think they pay for the bandwidth, electricity, equipment, backup tapes, diskspace and maintenance labour you're useing on their server?
Felching schoolteachers?
Writers imply. Readers infer.
I see some people posting some negative replies, or lamenting on OSS process, etc and saying how it didnt work, or how a psuedo-trusted patch would get in, if it went through proper channels, or some such crap.
this couldn't be further from the truth, you are all forgetting many things, #1 - the checking scripts run daily now, and Larry has mentioned he's going to step that up, still fixed within 24hrs is a damn good response time! closed-source could never be this fast.
#2 - all this talk of peer review, saying it didn't catch this or whatever nonsense, yes in a way it did, and whats more it's exactly what will keep semi-valid attempts or those through "proper channels" out of the code. You forget, millions of people around the world review this stuff, and someone, somewhere will find it relatively quickly, and not just because all the good developers (which is most of the millions) really LIKE linux and do their utmost to protect it, and ensure that no twits do things like this.
on the oft-side billions to one chance someone does something stupid like people said hire someone to do good patches for a long time, get trusted, and submit a patch with this kind of code in it, well, first of all, this is just stupid, it would take years to get that trusted from "zero", second, even assuming all that, the code would still get caught very quickly.
Like I said, someone, somewhere is gonna notice real quick, because the millions of us out in the world really happen to LIKE linux, and protect the kernel most of all, and I'm sure as the code worked its way into the tree, one of the people would catch it, and I'd be willing to bet several would see it at the same moment, including Linus, et all.
You simply can't pull a fast one over the great coders we have, these aren't your average coders, and remember, not just them, but all of us, really, in a way, put our heart and soul into supporting Linux, its a confidence we dont share lightly, the kernel is the most protected of it all, yes, for obvious reasons, its the most critical code.
But even outside the kernel, remember millions of people around the world are reviewing code 24hrs a day, every day, and posting notes about issues, patches, etc.
It's simply much harder to get by all that. Like I said, and I'll say it again, someone's gonna notice, and probably LONG before it even gets into the main BK tree, because even those reviewers ain't slouches!
Closed source has a smaller review team, and I know for a fact internal developers add back-doors to code all the time. I know many closed-source coders (not necessarily personally) that as a matter of habit throw in back-doors into every piece of code they write, because they hate their job, and the people they work for, and hate the product. Since very few people ever review the code, things can sit there indefinately and never get found.
remember this is a work of pride, something the community really cares about, we really want to see it succeed, and not have the issues like this, or that others have, we want to protect it at all costs, in any way, to ensure a good future, and protect the users out there.
remember, we're users too! If it means that much to you, wouldn't you be checking it too? damn straight! This is exactly why the OSS model is so damn important,
and its exactly why Microcrap, SCO, etc will never "get" it. I'd even add Intel to this list, because I think AMD is really "getting" it.
summary - we like it, we care about it, and aint no way we gonna let some dork attempt to ruin something we've worked so damn hard to build, not just for ourselves, but for everyone, its a matter of pride.
and yes, anyone found out (and they will be!) doing this shit is gonna get their ass kicked into next week...
Microsoft insists the timing of their bounty (pay deal) on (for) virus writers (hackers) "entirely coincidental" (damned convenient)
Curiosity was framed. Ignorance killed the cat.
Actually Linus has an opinion:
On Wed, 5 Nov 2003, Andreas Dilger wrote:
>
> Granted that this was not a break in BK itself the event is still alarming.
> It makes me wonder if there is some way we can start using GPG signatures
> with BK itself so that you can get proof-positive that a CSET annotated
> as from davem really is from the David Miller we know and trust.
A few things do make the current system _fairly_ secure. One of them is
that if somebody were to actually access the BK trees directly, that would
be noticed immediately: when I push to the places I export from, the push
itself would fail due to having an unexpected changeset in the target that
I don't have on my local tree. So I'd notice that kind of stuff
immediately.
And that's likely to be true of all other BK users too: the public trees
are just replicas of the trees people actually _work_ on, so if the public
tree has something unexpected, trying to update them just won't work. You
just can't push to a tree that isn't a subset of what you already have.
So any BK corruption would have to come from the private trees, not the
public ones. Which tend to be better secured, exactly because they are
private (ie they don't have things like cvspserver etc public servers). I
suspect most of us have firewalls that just don't accept any incoming
connections - I know I do.
I think it's telling that it was the CVS tree and not the BK tree that
somebody tried to corrupt.
Linus
-- jaf
...is the primary CVS repository reachable? Nobody should be able to touch it, even just to read it, except the chosen few. Or am I mising something?
You're such a douchebag. A single banner ad that is a) an image, b) fixed size, and c) fixed placement on the top of the page is one of the most UNobtrusive ways of advertising possible. How'd you like popups, popunders, expanding "CLICK TO CLOSE" ads, ads with sounds, ads that try to install Gator on your PC, ads with JavaScript errors in them, etc...
/. doesn't get /.ed? Because it runs lots of hardware and bandwidth that's supported (partly at least) by money from those ads. How'd you like to pay for access? No? Then shut up and be appreciative that the ads are tastefully executed on this site, rather than trying to subvert the system that's effectively in place to give you something you like for free.
How do you think
You combine an essential failure to reason with a wonderfully concise lack of insight.
In truth it wasn't the nature of the two systems (BitKeeper vs CVS) but the very nature of an interface existing between two systems.
It is (verrily 8-) the programatic equivlant of many eyes in action.
Multiply that, your core error, by the straw-horse of communitizing the open-vs-closed nature of the two products.
It is the particular script that spanned the gap that "caught it".
But further still, the "many (human) eyes" never played in to the operation. Someone electronically installed a hack and it was caught by a script (which script may or may not have underwent peer review). That that same hack would have then had to face peer review only doubles the effectiveness the the kernel's management paradigm.
So "yea, it was caught by a script" yields NO POSSIBLE VALID INDICTMENT of open over closed source models.
Innocent people shouldn't be forced to pay for inferior software development.
--"Code Complete" Microsoft Press
It doesnt have to do with the development process.
Besides, the linux kernel is considered to be a monolithic kernel as well.
it wasn't SCO trying to sneak in some of it's code so it could point to it and say... "See? We TOLD you the kernel had our code in it!"
But seriously, it's a credit to the ppl working on the project that this attempt was caught. I am not toooooo worried that other attempt at malice have gone unseen as just too many people are looking over the code all the time. Any maliciousness would have to be buried under several layers of valid and useful code (not sluggish crap or slop, either).
After hearing of security holes in every OS AFTER the fact, it's nice to see one caught before it even happened. And a non-accidental one at that. Kudos to the team!
If SCO's case ever hits the courtroom, this incident might mean that it's impossible for them to claim anyone knowingly added SCO property to the Linux kernel.
For all we know, they could have been doing it themselves...
To Terminate, or not to Terminate, that's the question - SCSIROB
This is the *Linux kernel*- arguably the most watched open codebase in existence-
And there's questions about whether this is the first time?
IMO the positive "peer review" aspect of open source just doesn't trickle down to smaller projects.
I browse at +5 Flamebait- moderation for all or moderation for none.
This is why monolithic kernals, liek the OpenBSD kernel are bettar. Since Theo dee Raddt is the only one who can edit the code, he is the only one that can add or remove back doors and exploits, so this kind of thing would not happen
Uh, a 'monolithic kernel' is a kernel that handles a lot of stuff, not a kernel developed by only one person.
In any case, this code was added after the server hosting it was hacked. And no development model will save you from insertion of malicious code if you can't rely on the security of your host machines.
I'm so tired, for a second there, I thought you were dissing Burger King.
I believe several of the BSD's use CVS. I wonder if their code have been hit as well or is it just Linux that was targeted?
Might be a good time to check them to see if somebody decided to pull a similar exploit against other OSSs.
I prefer the "u" in honour as it seems to be missing these days.
Back-Door Hack
Was the password "Jeff"?
Anyone that thinks security is easy (apparently some people still do) really needs to read Ken Thompson's 1984 Turing Award Lecture "Reflections on Trusting Trust": http://www.acm.org/classics/sep95/ As Bruce Schneier says, security is a process, not a product.
but I trust brackets more than I would trust my own ability to remember the precedence of every operator in C.
I consider the parenthesis to be two character comments.
Ignoring ourselves, its also a matter of trusting others. I've seen people break perfectly good code by replacing a multiply by a power of two with a shift in a non-trivial expression not realizing the difference in precedence between multiply and shift. I've learned to write code others will have an easy time maintaining it so I am free to do other things that interest me more.
...some of you always start bagging MS.
VENI, VIDI, VICI, DIXI
If you change that to if((c=1));, the error will be supressed - the extra parenthesis show that the assignment was intentional. Which it was, in a way.
Although, since the variable is assigned a constant, this is a bit stupid (except for some weird overloading of =)
The fact that this was possible once, should really make people think about the possibility of it happened ALREADY, and determine if it is necessary to hunt through the code for a systematic review
Like some months ago when the FSF got root'ed and they spent a few days trying to figure out if any GNU source was compromised?
Before that didn't OpenSSH briefly have compromised source? My recollection is more fuzzy on this one.
And this is just high profile recent history. Imagine the stuff waiting for folks at SourceForge in smaller projects maintained by "less talented" people.
Hey, how about those hidden racist comments in Perl scripts?
&debug("[$.]: ${_} -> $ent->{$_}");
dark ebony black underground giraffes with some money inside and outside parentheses are sent to entertain for money while shooting arrows at rich quotation marks while winking?
I am Monkey, the Great Sage, equal of heaven!
In general I agree with your sentiment that readability is very important but I don't see a readability issue in this particular instance. When I first encountered (0 == n) rather than (n == 0) my thought was that's odd. Odd as in non-typical not odd as in perplexing. Is (n == 0) more readable or does it merely mimick the style of the textbooks we read?
Makes you wonder how many backdoors have slipped through in windows, in Linux anyone can discover a backdoor. Who discovers them in windows?
Well, I'm hearing a lot on here about "hey, if this happened to open source once, it may have already happened or it may happen again" ... sure... but wait a minute, wasn't the halflife 2 source code leaked/stolen a while ago? Propietary code, no? Who is to say that they didn't add a few dozen back doors into it while they were at it, mmm?
The risk of this happening to open and closed source is roughly the same, and the benefit of open source it that it can be spotted...
That one wouldn't get very far, now would it...
Both Visual C and GCC will not produce the warning if there are parenthesis around the assignement statement, as there are here.
What annoys me now is that some games developers (e.g. Rare) are being called "2nd party" because they are owned by a manufacturer.
The correct terms are: 1st party = manufacturer, 2nd party = end-user, 3rd party = anyone else.
The vandal who put this in the CVS code tree obviously had a lot of skill.
It's a clever backdoor, and might have gone unnoticed, if not for those those good automated checks in the BitKeeper-to-CVS gateway. Notice that the particular coding style is a common C gotcha (using "=", assignment, instead of "==", comparison). At first glance it looks like the value of uid is being compared with 0, when in actuality it is being assigned the value of 0: root! The gcc compiler is good about warning for this, except that this too has been defeated: as mentioned on the mailing list, notice the unusual high number of parenthesis around this expression. That high number of parenthesis has the effect of suppressing the gcc compiler warning.
So, whoever did this obviously knew what they were doing and tried to obfuscate it. As somebody else mentioned on the kernel mailing list, if somebody is going to put in a backdoor like this, why not make it a remote root hack?
As it is now, the above hack is only locally exploitable. A process on the local system still has to call the wait system call with that particular combination of flags, in order to trigger the exploit and get root. To my knowledge, no known applications do this, because the combination of flags is supposed to be invalid.
If a spammer or somebody else was trying to backdoor the Linux kernel in order to gain a large number of machines to infest, then one wonders why they didn't put in a remote root exploit. It seems strange to go to all the trouble. Since this backdoor attempt has been caught and blocked, security will now only become tighter, and they might not ever get another chance like this.
Maybe it was intended to be used with another application, also backdoored in the same manner? It might be insightful to scan other open source applications and search for this particular usage of flags to the wait system call.
In any case, I'm glad this hack was caught!
Dr. Demento On The 'Net!
So, if source code for a trojan is inserted into a GPL's project and assuming the author of the Trojan knows the trojaned program is a GPL project, does that mean that the Trojan is technically GPL'ed
I'd rather have a kernel with lots of (known) submitters than a single psychopath.
Cole's Law: Thinly sliced cabbage
Or.. Should that be:
ALIANWARE == OVERRATED
It sure makes quite a difference.
Does anyone remember that choice item, which was used as the password for a backdoor coded into asp some years ago by a witty Microsoft engineer? I was sitting next to a friend of mine as he cracked a local online store with this vulnerability to see if it actually worked. He was a decent guy, Linux fan through and through and informed the company about the hole.
But what if he hadn't been?
it was a computer error, computer error.
weapons of mass destruction.
we flunked.
Heh, I remember those old monolith demos. funky loops.
Should end the old argument about which is better, the habitual, easy to read, but easy to screw up or abuse:
if(variable == CONSTANT) { }
Or the safe version that's so much harder to screw up and which turns out to be just as easy to read with practice:
if(CONSTANT == variable) { }
Do we all understand the real world significance of this now?
If you still want to advocate (variable == CONSTANT), then please feel free to prove that no accidental or abusive (variable = CONSTANT) exist in the kernel.
If you were blocking sigs, you wouldn't have to read this.
I recommend reading the original mailing list postings, they're good reading, rather better than most of these slashdot comments. Just follow the first link in the article.
There have been several now.
Looks bad? Think about it.
Each one I've heard of has been detected, corrected and publicised within hours. We don't know about any that haven't been found, of course, but I haven't known of one that went undetected for even a day. If it was easy to slip in a deliberate security hole that escaped initial detection we would hear about holes that were detected only after months or more.
This has been a test of the proposition that all those eyeballs don't realy improve security because most don't realy read the source. The proposition has failed.
All those eyeballs make for an almost certanty that at least one will take a critical look.
That is all it takes.
"Uh, a 'monolithic kernel' is a kernel that handles a lot of stuff, not a kernel developed by only one person"
Uh, no it's not.
"In any case, this code was added after the server hosting it was hacked. And no development model will save you from insertion of malicious code if you can't rely on the security of your host machines."
True, but I don't see Linus' tree being compromised. Linus even said that he'd have spotted these changes as they tried to be
updated onto his machine.
This was a belt, braces and safety-pins setup, and only the safety-pins broke.
Phil
Your head of state is a corrupt weasel, I hope you're happy.
oh bugger! I snuck it into this other o/s ok, why can't I hide it in linux?!
those kids should look whom they are messing with. In fact, they are messing with the most l33t people available in the world right now.
Let's see what comes next %)
Which dimensions are, in fact, easily explored. Simply take
this little pill => .
Get thee glass eyes, and, like a scurvy politician, seem to see things thou dost not.--King Lear
Maybe someone should grep the source for "uid[ \t]*=[ \t]*0[^0-9]" and inspect any matches, just in case?
#naabhaprzrag, #sverubfr-000, #agi-fcbafberq, negvpyr[pynff*=' negvpyr-ary-'] { qvfcynl: abar !vzcbegnag; }
If a spammer or somebody else was trying to backdoor the Linux kernel in order to gain a large number of machines to infest, then one wonders why they didn't put in a remote root exploit.
We saw the code that was used to gain root access locally on the machine. Perhaps the bit of code (buried elsewhere in the source) that allowed for remote access as a normal user has yet to be found, no?
Of course, at some point, we do have to trust someone.
Ken Thompson wrote an original speculative essay on this for CACM back in 1984 of all years.
It is really well worth the read. The short form is that there exists a way to subvert the compiler such that it is no longer trustable and it will build a back door into the OS forevermore. This paper is a must read.
Given that I don't understand the code in the slightest (well I understand the assignment and user id 0, but not what __WCLONE or __WALL does or relates to) how would a standard user have been able to get him/herself root?
Avantslash - View Slashdot cleanly on your mobile phone.
Turns out you need an ACM subscriptiong to the link to the PDF above.
Ken Thompson has been kind enough to post a link to a free online version in the classics library at the ACM.
An old trick, from my days writing with C compilers under MS-DOS: to force the compiler to catch this problem, put the constant (the zero, in this case) on the left-hand side of the 'equals' sign. C doesn't allow assignments to constants, and even a crude compiler will catch this.
In other words, when you meant
and used a single 'equals' sign, instead of you'll getI bet you don't have a girlfriend :P
They screwed up the Matrix, now they're trying to screw up Linux too!
Squares of the first three primes...
"...and why did we stop at three dimensions?..."
Read the book.
The living have better things to do than to continue hating the dead.
This dispels a lot of the FUD surrounding whether Linux kernel maintainers can be relied upon to detect such false entries.
In a lot of ways, this is a Good Thing, in that it proves, conclusively, that Linux' QA works and works damn well.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
Plenty, probably, but that's irrelevent anyway. This person wasn't authorized to put anything in, and there are plenty of laws against unauthorized "hacking."
I would imagine putting a backdoor in an OS project is legally quite the same as in a proprietary one.
Sticking feathers up your butt does not make you a chicken - Tyler Durden
Uh, no it's not.
No seriously it is.
A microkernel is more modular but is slower I believe due to the huge amount of message passing required for the modules to communicate.
Chris
My God, it's full of source!
Escher was the first MC and Giger invented the HR department.
> Squares of the first three primes...
I thought it was squares of all positive integers. After all, is 1 really, really a prime number?
--
-JC
The only way to get the code to execute was to actively try to break the copy protection on Word 2.0. If you were running under a debugger and had patched out the first two of the three anti-debugger checks, that message would be printed out, and Word would start randomly reading data off the disk.
Peter Norton found it by looking at the word binaries on the disk and put it into his column, and Microsoft had to pull it immediately.
Needless to say the intern involved never came back to Microsoft.
So what's your point? That consumer operating systems have to ship with the default logon as root?
Don't you think that "does lots of stuff" and "runs a single process" are different things?
I do lots of stuff, but I'm not a monolithinc kernel, for example.
Do you have any books, references, or lecture notes to justify the view that "does lots of stuff" means "runs as a single process"? Thought not.
So seriously - no it is not.
And microkernels can theoretically be faster in SMP situations, as the different kernel services can run independently with uniprocessor affinities. But get the affinities wrong and you can make the SMP as bad as a uniprocessor microkernel setup and worse than any monolithic setup due to the reasons you state.
I just hink that if people are going to try to give explanations that the explanations should be meaningful. I feel "does loads of stuff" is a gross miscategorisation of what monolithic kernels are.
The fact that the post was trying to correct something that was way way way more bizarre didn't help. It was 100 times nearer the mark than the "one author" implication, but still missed, IMNSHO.
YAW.
Your head of state is a corrupt weasel, I hope you're happy.
many years ago when easter eggs in programs were all the rage, there was this one guy who programmed an easter egg with a guy mooning you from inside a VW bus.
From excellent karma to terible karma with a single +5 funny post...
No BitMover infrastructure was compromised, the machine in question is a public machine maintained by the kernel developers.
I think your anti-BK colors are showing when you can take a situation where we prevented a trojan horse from being added to the kernel and your reaction is that we screwed up.
It's in the LKML archives, read the thread there.
Also, a microkernel has the potential to be much faster than it is today, if hardware were setup for it a bit better. Most(all?) modern CPUs have 1 priority bit, that basically determines if the system is running in a system or user context(note that this has nothing to do with root/user permissions.)
If a CPU had more than one priority bit, the various system daemons/programs that microkernels use could be each given one priority bit(or a combination of them, etc., but that's minor.) As it is today with a system like HURD, the microkernel emulates those multiple priority bits.
Note that with a microkernel(more for the grandparent poster) that this means that the kernel itself can be quite tiny, and drivers can be regular old programs(almost, not quite regular, but close.) Which means, if one of your drivers-- say your network card driver-- gives out, it can just be restarted by itself.
All the kernel has to do is pass messages back and forth, and coordinate the other programs. (IIRC, even the scheduler can be out of the kernel proper)
If I was that drunk, I would have remembered it -- H. Simpson
If RMS wants to rant about revision control systems, he'll need to say that CVS needs to be replaced with a more functional alternative (Subversion, perhaps), not BK.
Or Monotone, perhaps.
Monotone is a new revision control system being developed by Graydon Hoare at RedHat. It's notable for having cryptographic hashs and signatures implemented through the entire system.. each delta in the archive has a signature associated with it, as does each bit of information about the delta.
I'm not sure how well such a system would perform, but there's no sneaking data into a system like that without subverting (sorry) someone's private GPG key.
- jon
Ganymede, a GPL'ed metadirectory for UNIX
or do i smell the stench of M$ somehow on this. A) you'd have to be pretty skilled to know even how to code for the kernel and to exploit it in this way. We may take this knowledge for granted here on /., but cmon people, the guy needed "5K1llz". B) Then, he needed a motive. What better way to keep your product on top than to prove that your enemy is riddled with a security hole worse than the the Nimbda worm?
This would most likely be a "test" run, since they would be stupid to think that no one would pick this up. They just wanted to see if it was possible. Now that it is, we'd better watch out. They will be subtle next time.
I shut the machine down so that we can reinstall it.
cvs.kernel.org is in NO WAY the "primary site for the linux kernel", it's a BK/CVS/SVN server that BitMover and Penguin Computing provided to the kernel developers.
The primary BitKeeper site is up and functioning and even if it wasn't, that's no problem because BK is replicated. Unlike CVS/SVN you'd have to find every single replica and kill that to shut down all the BK trees. It's impossible, there are 10's of thousands of replicas.
This is just a CVS problem, inspite of the noise on the kernel list the CVS server is very lightly used, maybe 3-4 people, so it's just not a big deal. The trojan horse is a big deal but the fact that it got into CVS is fairly minor and it was caught by us immediately.
Clearly, the best way to get a remote backdoor is by making two changes, each of which is insufficient to garner attention but which, taken together, constitute the entire exploit.
The complementary step would be to try to get the priviledge elevator installed in some system component so that together they could be used as a remote bakdoor when a vulnerable kernel was running a compromised service.
Time to examine recent paches to common daemons like Apache, Samba and ftpd to see if there's anything which might take advantage of such a hole.
I have discovered a truly marvelous sig, unfortunately the sig limit is too small to contain i
Why use a back door, when the Windows is open?
I wonder if there is not a way to defeat such a method?
For those who didn't read the article by Ken Thompson ( read it here) a compiler is corrupted so that it inserts a bug into all compilers that it compiles, and the purpose of that bug is to insert a bug into another program (such as login) when it compiles it (such as accepting a certain password as the root password)
Both bugs have to be a pattern based search method. They look for some string or some sequence of characters that the original hacker believes will be consistent in future code, and then make their modifications.
Running the code through a obfuscating precompiler that both randomized variable names and added random white space would potentially remove any pattern that the trojan was looking for.
Can anyone think of things that I missed (or ways to make the trojan continue to work in the face of obfuscation)
the obfuscator would, of course, be written in an interpreted language... ( [raises pink to corner of mouth and channels Dr. Evil] whose interpreter has of course been corrupted so that it inserts naughty limericks into every application it "obfuscates".... MUWAHAHAHA... MUWAHAHAH....)
--
Was it the sheep climbing onto the altar, or the cattle lowing to be slain,
or the Son of God hanging dead and bloodied on a cross that told me this was a world condemned, but loved and bought with blood.
...not only are you a troll, you can't even make up your own stuff, you have to use someone else's.
PHEM - party like it's 1997-2003!
I couldn't help thinking that if this kind of attack was done within the closed-source BitKeeper instead, and we didn't have open sourced CVS to ferrit it out - it might never have been caught.
First off, white space is removed in the lexical analyzer in the earliest phase of compilation, so that part is way off base. The pattern matching would occur on something like expression trees (parse trees, syntax trees, intermediate code tuples, ...)
Secondly, the part being matched might not be alterable. If the target code is doing something like "strcmp(test_password, stored_password)", then you can't obfuscate the string "strcmp"; the symbol wouldn't be resolved by the linker.
Thirdly, the standard obfuscation tool would be well known, and the compiler could be adapted to it, or it could be subverted as well.
Fourthly, the real point of Thompson's trick was that he got people to use a trusted binary of a compiler to recompile the world. If people were distrustful enough to want to run an obfuscator, they would also refuse to trust the binary compiler sent to them by that rogue Thompson, and instead would use one of their own.
You can try to get clever, but there's no avoiding Thompson's ultimate point: there's always someone or something that must be trusted, and that is a weak link.
After all, accidental bugs that cause security flaws show the impossibility of even completely trusting oneself.
Professional Wild-Eyed Visionary
Whose university is wheening off Sun boxes. Either you have desktop machine, then you already have root access or you have an account on a corporate server, then you are probably supposed to manage it. You can go and hack your own ISP, but these are the people in the best position to monitor you and they know exactly which billing address to show up at and kick your ass *shudder*.
Why not hack NFS code instead - or better yet SSH, Apache, gcc and so on? This exploit is so ineffective that I wonder if it's a bug after all.
Couldn't someone just slip nefarious code into the source they are hosting on their mirror, then compute new hashes so that everything appears legit? You could also rebuild the binaries that you are hosting to include this evil code. And even if you are not mirroring, the GPL lets you create your own distro as long as you publish source. Just add the nasty code, publish it, and trust that none of the users will go hunting for problems. After all, it was the source MAINTAINERS who found this particular attempted exploit, thanks to some tools they were running.
I'm certainly no expert on Linux mirrors and the like. Does anyone know if there is a way to prevent what I describe?
One obvious advantage to a microkernel is that portions of the kernel can then be (carefully) swapped out to virtual memory on disk, whereas monolithic kernels generally cannot (just don't page out the pager, or disk I/O system!). This is good, because those microkernels tend to run significantly larger than a comparable monolithic kernel. Microkernels tend to be handle to handle real-time scheduling constraints better than a monolithic kernel can because of the nature of the thing.
Now, aside from SCO UNIX products, I don't know of any OS that is completely monolithic these days. Linux has kernel threads and run-time installable modules which are rather un-monolithic from a purist stand point. And from the microkernel camp, we have NT and OSX, which runs their entire kernel in the highest priority level for performance reasons, and in NT's case, have a rather heavy microkernel. Hybrids, the whole lot of 'em.
I used up all my sick days, so I'm calling in dead.
You exaggerate a lot; obviously you have never worked in industry.
No, I exaggerate a little and have worked in the industry for 10 years as a code monkey. I've been up the totem pole from the bottom to the top - programmer, lead programmer, project manager, and IT director.
When I was a peon, I watched marketing take over development and make a lot of numbskull decisions not unlike the ones I described in my last post. I've worked with companies where company policy forbid them from making simple no-brainer decisions just like the example we're discussing.
If you haven't seen crazy stuff like this, then I doubt you've worked in the industry very long.
(n == 0) doesn't mimick the textbooks, it mimicks the natural language. "If n is zero..." (That's English, but I don't know of any language where the normal order is reversed, except for Yoda-speak.) That's what makes it more readable.
..." is ((n == 0) || (n == 1)). I understand your point, when originally learning to program I tried using (n == (0 || 1)). I think your argument may have more merit when writing an introductory programming textbook. However once we get to a level where a programmer has a non-trivial amount of experience I think this argument falls away.
Similarity to natural languages does not seem to be a very good argument since programming language are often inherently different. For example "If n is zero or one
It's the quiet ones you've got to watch...
While you're watching the quiet ones, a loud one will fucking kill you! -- George Carlin
boldly going forward, 'cause we can't find reverse
No, I think he means monotheist
No BitMover infrastructure was compromised, the machine in question is a public machine maintained by the kernel developers.
Oh, then I'm sorry for the false statement. So the kernel developers are unable to guard their own sources. Still scary, but in a different way.
This idea was invented by Shampoo.
wouldn't it be good to include the users gpg pki sig with the code they send in, and if linus or anyone else modified it, they would sign their diffs or the code itself? Not inline, but a seperate .asc
Thank you for addressing my post. Let me respond to your points.
1) While I did feel a tiny bit bad about it, I called the poster a douchebag not because I don't agree with him, but because he is spreading harmful information that has the potential to negatively effect all of our ability to access this site for free and via the high bandwidth connection it uses (off-topic in a discussion about a Linux kernel back-door, by the way). I'm all for open discussions without name-calling, but this went beyond creationism/evolutionism to something more akin to "how to avoid speed traps near schools" which are in place to make the world a safer/better place.
2) First of all, saying "people don't actually LOOK at banners" is just plain ridiculous. Some people do, and some people don't. With TV, some people watch commercials, and some go make popcorn and take a leak off their roof. Text ads may generate more click-through, but they are more insidious in that they begin to merge with the content, something a clearly demarcated graphic ad does not do. Your evidence does nothing to support your claim that abolition of graphical banner ads would be of mutual benefit to anybody, not that you've even defined "benefit", and if you consider the issue I just raised about text ads, I can't imagine how you could make that claim.
3) I never made the claim that banner ads and Slashdot are in some kind of eternal embrace that we can't shatter. I pointed out that "the ads are tastefully executed on this site" and that you shouldn't publicly try to "subvert the system that's effectively in place to give you something you like for free."
P.S. I am posting this without Karma Bonus because I don't need to further publicize this (nor encourage wasting of mod points). I just wanted to leave a response to your issues attached to this thread.
didn't microsoft release Back Orifice
Well, I guess that means all the closed source developers have the same problem. And I guess they probably don't know either. Just wondering, if open source software facilitates somewhat easier detection of subversions, are undetected subversions in OSS, also not at a greater risk of being exploited by hackers, especially because the code is in plain view to a larger audience? (as compared to a much smaller closed source development team..)
If one wants to put on the conspiracy hat - and with this example, that doesn't seem too unreasonable, since *someone* was trying to plan a backdoor - this is a very clear warning that the security of the Linux source code needs to be taken very, very seriously.
If - just hypothetically - some Huge Opeerating System Seller - really wanted to discredit not just Linux but the whole Open Source method, what better way than to plant something like this and then step back and say "Look! Open Source is insecure! Why, with all those strange foreign bohemian types working on it, who knows *what* one of them might slip in?"
Yes, you and I know all the holes in that argument, but the pointy-haired types wouldn't.
(When properly configured).
:)
At least gcc 3.3.2 (what I tested on), yields a: "warning: will never be executed" warning if compiled with "-Wunreachable-code" (in addition to -Wall, which does not really add "all" warnings).
This is because "&& (uid = 0)" always yields false, and the "then" part of the if never runs.
Maybe its a good idea to compile OpenSource code with all these warnings turned on
I mean, at some point someone might try recompiling the compiler from the begining. And eventualy someone might simply change the compiler code so much that the back door no longer works.
autopr0n is like, down and stuff.
If they arn't running as root, they won't let you do what they want. Unless there is a local exploit....
autopr0n is like, down and stuff.
"Fairly secure" != "Secure"
.Net technology... or maybe not. :P
Unless you're using some new
Yes, by whoever put it in.
Litigious bastards
Well, this was posted to cnet news.com: Attempted attack on Linux kernel foiled By Robert Lemos, CNET News.com Friday, November 7 2003 10:07 AM An unknown intruder attempted to insert a Trojan horse program into the code of the next version of the Linux kernel, stored at a publicly accessible database. Security features of the source-code repository, known as BitKeeper, detected the illicit change within 24 hours, and the public database was shut down, a key developer said Thursday. The public database was used only to provide the latest beta, or test version, of the Linux kernel to users of the Concurrent Versions System (CVS), a program designed to manage source code. The changes, which would have introduced a security flaw to the kernel, never became a part of the Linux code and, thus, were never a threat, said Larry McVoy, founder of software company BitMover and primary architect of the source code database BitKeeper. "This never got close to the development tree," he said. "BitKeeper is really paranoid about integrity, and it turns out that was key to finding this Trojan horse." (Emphasis mine) Linus Torvalds, the original creator of Linux and the lead developer of the kernel, uses BitKeeper to keep track of changes in the core software for the operating system. On a daily basis, the software exports those changes to public and private databases other developers use. An intruder apparently compromised one server earlier, and the attacker used his access to make a small change to one of the source code files, McVoy said. The change created a flaw that could have elevated a person's privileges on any Linux machine that runs a kernel compiled with the modified source code. However, only developers who used that database were affected--and only during a 24-hour period, he added. "The first thing we did was fix the difference," he said. "It took me five minutes to find the change." When BitKeeper exports the source code to other servers, it checks the integrity of every file, matching a digital fingerprint of its official version of the file with the version on the remote machine. That comparison caught the change to the code stored on the server. The changes looked like they were made by another developer, but that programmer said he hadn't submitted them, McVoy said. The recent incident raises questions about the security of open-source development methods, particularly how well a development team can guarantee that any changes are not introducing intentional security flaws. While Microsoft code has had similar problems, closed development is widely considered to be harder to exploit in that way. Linus Torvalds addressed the issue in a post to the Linux kernel mailing list. "A few things do make the current system fairly secure," he stated. "One of them is that if somebody were to actually access the (BitKeeper) trees (software repositories) directly, that would be noticed immediately." A critical security flaw was found in CVS in January, but it's unknown whether the attacker used the vulnerability to gain access to the CVS database. BitKeeper's McVoy hopes the current incident will quash objections raised by some members of the development who don't want to add a new feature that would require all changes to be digitally signed. Even so, he said, the open-source development model likely would have quickly turned up any security flaws. "A Trojan horse is just a bug that a person has put into the system deliberately," he said. "The open-source security model is that everyone is using this stuff, so bugs get found and get fixed. That's one of the reasons that you are not hearing me freak about this." McVoy said the disk from the compromised server has been saved for later analysis, but any decision to contact law enforcement belongs to Torvalds and others. Torvalds could not be immediately reached for comment. http://asia.cnet.com/newstech/security/0,39001150, 39156944,00.htm
I stand corrected. I didn't think any hardware around today actually had that. (mostly have only messed with sparcs at that low of a level.) Now it makes me think I should try Hurd again sometime soon:)
Thanks for the info.
If I was that drunk, I would have remembered it -- H. Simpson
nothing like a good kernel vulnerability to bring out all those low UID posters!
-R'
The strings that such a compiler are looking for aren't C keywords or macros, so the preprocessor won't change anything. Identifiers can't be broken up without breaking the program, so you can't add random white space in the middle of words. Adding random white space in between words has never affected the C family of languages (if it did; there wouldn't be any holy wars over indentation styles, because only one method would work, and we'd all have to use it.)
At some point, an implementation of login(1) has to look at "/etc/shells" and call the user ID resolver routines and drop privs and exec what will probably be one of three or four known shells. If your compiler hack detects "/etc/shells" as a string literal, then proceed to the next test.
You cannot apply a technological solution to a sociological problem. (Edwards' Law)