Slashdot Mirror
The Internet's Biggest Security Hole Revealed
At DEFCON, Tony Kapela and Alex Pilosov demonstrated a drastic weakness in the Internet's infrastructure that had long been rumored, but wasn't believed practical. They showed how to hijack BGP (the border gateway protocol) in order to eavesdrop on Net traffic in a way that wouldn't be simple to detect. Quoting: "'It's at least as big an issue as the DNS issue, if not bigger,' said Peiter 'Mudge' Zatko, noted computer security expert and former member of the L0pht hacking group, who testified to Congress in 1998 that he could bring down the internet in 30 minutes using a similar BGP attack, and disclosed privately to government agents how BGP could also be exploited to eavesdrop. 'I went around screaming my head about this about ten or twelve years ago... We described this to intelligence agencies and to the National Security Council, in detail.' The man-in-the-middle attack exploits BGP to fool routers into re-directing data to an eavesdropper's network." Here's the PDF of Kapela and Pilosov's presentation.
Must have the world's largest collection of online porn.
Which would figure, actually.
Help stamp out iliturcy.
Everyone loves sensationalist news headlines. *sigh*
Anyone have any insight as to how serious this ACTUALLY is?
I hope that all of those people who thought that getting users to blindly accept self signed certs was a good idea are starting to feel a bit stupid now...
An SSL cert signed by a trusted central authority isn't the absolute solution to all mitm attacks, but it's a whole lot closer to 'safer' than not.
I find the thought of this genuinley scary. Correct me if I am wrong, but we would have to change the BGP protocol itself to fix this issue. That isn't going to happen anytime soon I reckon, so I guess there is nothing we can do but encrypt senstive transmissions and hope for the best.
BGP is almost always setup manually, at least when first configured. Network admins: DO NOT PUT UNTRUSTED PEERS IN THE ACLs. Joe smith running BGP on 123abcxxxhost.nl has no business being in your tables. If you're accepting adverts from any AS you deserve what you get.
The routing on the Internet has always been hierarchical: get updates from your upstreams. If they send you bad info you're SOL anyway, just like SSL certs and Verisign's root certs.
Website Hosting
...that the good folks at the NSA (and/or the FBI, CIA, DHS, ATF, etc., as well as their counterparts in other nations) have been exploiting this for years.
With spending like this, exactly what are "conservatives" conserving?
This is the guy who taught the NSA how to spy on us en masse.
I'm glad he exposed the truth. Now we can protect against it. right?
They're using their grammar skills there.
A hacker marauding by the name "Goatse" exposed it quite effectively some years back.
I record my sleeptalking
OK, So do it. Now.
Really.
(Just don't wait until I am 90 and on Internet based life-support, without my consent because my money-grubbing heirs are just waiting for such a thing to happen then :-)
This issue is a bit more complicated than you think.
It was really cool, opened a lot of peoples eyes. Here is the archive, http://www.stits.org/fp/Defcon_16/. Please don't flood it and only download it if you will use the info. I also took a ton of photos: http://www.flickr.com/photos/stits/sets/72157606608859399/ Hope to see you all next year!
Wait, you're telling me that they taught US intelligence agencies and the National Security guys how to attack the internet with man-in-the-middle attacks and exploits to fool routers into re-directing data to an eavesdropper's network...
and they didn't do anything to end the interception and eavesdropping problem???
I am shocked.
-
- - You can't take something off the Internet! That's like trying to take pee out of a swimming pool.
The guy's been involved in many of security's moments in history.
Just disrupt the deflector shield with a tachyon burst.
... testified to Congress... disclosed privately to government agents... described this to intelligence agencies and to the National Security Council
So in other words, the US government knows about the issue. This is the United States government, people! Obviously there is nothing to worry about. Like, come on, as if the US government would allow eavesdropping on the information highways to even be possible. Like come on, srsly.
There is a lot of harm you can do, least for a short while. But I have to say, this seems like a lot of FUD to me.
It is not trivial to get BGP peering, or to keep it if you are doing bad things. You will need one or more peers, and they will have to do this for you manually, not automatically. And (as I can attest) the AS prepending this attack relies on is a very blunt instrument.
Here are the troubles I see
- You need to be able to offer a better path from Point A to Point B than the existing Internet topology
- Unless you are Dr. Evil and can afford infinite bandwidth, this better path had better not also apply to a large chunk of the Internet, or you will get hosed with a lot of bandwidth (and, also, instantly stick up on the screens of NOCs all over the place) and
- If you are relying on AS prepends, these affect the path from you, but not directly the path to you. They are notoriously tricky and may stop working (because of changes in other people's advertisements) at any time.
So, to me, this is a might work sometimes for some people in some places, but probably not that well on a general basis.
The DNS cache poisoning sounds a lot worse, frankly.
'I went around screaming my head about this about ten or twelve years ago... We described this to intelligence agencies and to the National Security Council, in detail.'
For a hacker he's pretty dumb. Everyone knows that the best way get attention directed to an exploit is to publish the entire kiddie-porn-folder of the person who can fix it, using the exploit in question.
If you quote this signature there'll be 72 copies of Windows ME waiting for you in Heaven.
I looked at this problem back in the early 1980s, when I was doing some work on TCP. I was trying to come up with a routing protocol that didn't require passing the same information around repeatedly, because backbone networks had very low bandwidth back then, and the existing routing protocols had either O(N^2) traffic or the "hop count to infinity" problem.
I came up with something called "Gateway Database Protocol", which was a scheme for passing tuples of the form "X says Y=Z" around. The idea was that any node seeing inconsistencies in "X says ..." would propagate the tuple back to X, revealing the problem to X.
This is enough to detect hijacking, but not enough to stop it. I'd worked out a scheme good enough to automatically correct erroneous data, but not one good enough to deal with the insertion of hostile data. The design goal back then was to guarantee that if the hostile site was removed from the network (perhaps forcibly), the system would then stabilize into a valid state.
That's not enough any more. But it is worthwhile considering that a routing protocol should have the property that if X's info is being faked anywhere in the network, X hears about it. BGP doesn't do that.
Instead he chooses to reveal the exploit to the NSA.
Let me guess. Next he'll find Osama Bin Laden, and then tell everyone using youtube.
If you quote this signature there'll be 72 copies of Windows ME waiting for you in Heaven.
I've seen implementations of ISIS, and have deployed it myself in both IP and ATM environments. I've never seen an actual deployment of ESES, and I've never heard of one either. I've encountered ISIS adjacencies which don't form correctly, and come up as ESIS, though.
What hardware supports ESES?
Need Geek Rock? Try The Franchise!
The whole MITM thing would raise a flag unless the attackers were close enough to the real routers for the ip address block it was hijacking. Several companies I know notice when BGP screws up and doubles their latency. They notice and complain loudly.
Well.. maybe. Or Maybe not. But Definitely not sort of.
Not quite.
Prepends affect your outbound announcements, and this affects inbound traffic to you. Prepends are the most effective tool for BGP manipulation because they're transitive - announcing more specifics works too, but that's not quite the same thing.
Need Geek Rock? Try The Franchise!
Great, give the very people who want to abuse this the most the inside details, then show shock when it isn't fixed.
I'm an American. I love this country and the freedoms that we used to have.
He gives the information to "national security" 10 or 12 years ago, and we only find out about it now!!!
Thanks for nothing, guy! I am sure the NSA had a real heyday using this information to spy on us without our knowledge.
Should we lynch him? Or just refuse to employ him because of his lack of judgment?
the problem with such a man in the middle attack is you are almost assured of being caught. unless you are sitting in the same complex as a backbone link someone is going to notice the huge spike in network laytency and track down where it's comming from. also since your inserting yourself between peers, it's like painting a giant target on yourself, similar to the target bubba from C lock is going to paint on your buttocks after the feds throw you in jail....
If you mod me down, I will become more powerful than you can imagine....
Didn't one Pakistan ISP rather graphically demonstrate problems with BGP when they null-routed YouTube worldwide?
We described this to intelligence agencies and to the National Security Council, in detail.
So this is how NSA is spying on me huh?
Enterprising hacker hijacks BGP and Rickrolls the whole world in 3... 2... 1...
You mean the user right?
The days of the digital watch are numbered.
You know, every day it seems there is another article about some other exploit discovered. Given the fact that DRM has been demonstrated to be doomed, I think we are seeing that basically all security is doomed. I think we truly are on the cusp of zero privacy. Basically we are at the point now where if someone wants to know about your electronic data, they can do it.
A work that expires before its copyright never enters the public domain and thus enjoys eternal copyright protection.
that requires one teensy weensy detail to work (in other words, one huge wonking detail)
here, it is to be a bgp level peer
kind of like i can empty a bank of all of its money
all i need is the key to the safe
yeah, minor detail
so do i panic now?
intellectual property law is philosophically incoherent. it is your moral duty to ignore it or sabotage it
So you told the NSA, the NSC and Congress all about this and they listened intently then sent you back to your lair/playpen/d&d fest whilst they began setting up MITM listening networks, and you did it FOR FREE. I'm sure they are eternally grateful for all you've done to make monitoring us that much easier. At least it works for everyone, so if you're not eavesdropping it's your own fault.
As my friend AJ used to say,"I'll work for $5 an hour, just let me take the trash out once a day".
A computer once beat me at chess, but it was no match for me at kick boxing. Emo Philips
So how is this so groundbreaking? IP prefix interception has been studied and discussed already. For instance:
A study of prefix hijacking and interception in the internet
from Sigcomm 2007.
Seems to be a much better work than this Defcon presentation.
Do not anger the worm.
yeah... fud shut get ipv6 up and running in no time... go on... =)
The US Government is using this exploit to spy on everyone.
So these guys go and convince the spooks that the Internet can be hijacked for comprehensive but totally stealth eavesdropping. And the spooks "don't do anything about it".
Except they do, don't they. The spooks go ahead and snoop the entire Internet. For the last 10 years.
I'm surprised at only the fact that the L0pht guys and others are still alive and running around loose to tell anyone that the spooks have known how to do this for this whole time.
Why is it taking so long for all Internet traffic to be encrypted end to end by default?
--
make install -not war
You called? Sorry I'm late
The Internet's Biggest Hole Revealed at http://goatse.cz/
> So Firefox's solution has been make it hard to pick the unsafe choice.
Except they really haven't. They've made it hard to make the sorta-kinda-theoretically-less-safe choice, the one that might result in a MITM attack, but in doing so they discourage SSL use generally.
Do you think that hypothetical user you're talking about is going to notice whether the page is using SSL or not? I doubt it. And a lot of companies seem to agree, and use plain old HTTP for all sorts of stuff when they shouldn't (we just had an FPP on this a few days ago, in fact).
As script-kiddyable as MITM attacks may get, they're never going to be as easy as just sniffing unencrypted traffic, and any time you make encryption difficult or complicated, that's the alternative people use.
"Ladies and gentlemen, my killbot features Lotus Notes and a machine gun. It is the finest available."
Some bgp attacks of similar nature have been shown in simulation. This is a paragraph from a related research paper: "This attack can be viewed as a variation of the well-known man in the middle (MITM) attack, in which players are ASes and messages are intercepted in one direction instead of both directions. Furthermore, it is more powerful than the MITM attack in the sense that it can affect traffic not just between two players, e.g. Alice and Bob, but between a number of sender ASes and one receiver AS, where each of the involved ASes bears a large number of end users. The impact of the false announcements made by the compromised AS depends on the topological properties of the compromised AS and the victim AS. Intuitively, if the compromised AS is located near the core of the AS topology it will affect more ASes. Also if the victim AS is located at the periphery of the AS topology it is more vulnerable to an attack." The full paper can be found in http://www.informs-sim.org/wsc04papers/038.pdf
Oh great idea, lets go straight to the NSA, FBI, CIA, SS and any other agency out there and explain in full detail how to spy on the entire world. Wow, real shocker they didn't fix this one. Even bigger Internet Security Hole: Best Intentions.
I agree, anything that can hijack a major protocol is seriously a problem, and the fact that they are demonstrating how to do it is even worse. The only good thing I can see is that I can't think of any REAL nerds who would WANT to take down the internet.
End-to-end encryption prevents eavesdropping.
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
*cough*Bullshit*/cough*
Read up a bit on PKI. Hint: The CA only signs your public key. When you request a cert, you do NOT give your private key to the CA. It wouldn't be a secret if you did that. And since the CA never sees the private key from the cert request, they don't have anything to give away except their own private key for the CA itself, which they're sure as hell going to keep secret.
Thank you!
European Solar Engineering School?
Electrical Status Epilepticus during Sleep?
Event-Space Exploration Strategies?
The issue of routing table poisoning isn't exactly previously unknown, and I guess quite a few have at some point suffered from problems with incorrect routing tables, which is essentially the same thing, though unintentional. There are techniques for limiting the problem, and I believe the "wasn't believed practical" to a rather high extent still holds for real purposes. I may be proven wrong.
Where's the news actually? BGP is out there for some time already.
Route poisoning can happen anyhow no big deal in BGPland unless holding very strict and extremely resource consuming receiving policies, which is plain unaffordable with 1500000+ announcements at any IX of your choice, changing quick in hard times.
BGP and the entire network infrastructure come from times of 'we can trust each other' innocence, the base concept itself is flawed and hard to change entirely in 'cost contain' times.
Not only (any) government agencies maybe in their right, but whoever is smart enough to get a peer in an IX: can be done.
Watch the AS of your traceroutes, if they aren't (yet) blocked by your ISP. Otherwise use massive encryption wherever possible and pray for the segment without encryption. Or trust people not to be all evil, common sense also an option.
RTFA too: the guys propose SBGP, bgp with signatures, I bet JunOS easily and IOS could afford it with stronger CPUs.
Conclusion: was a nice marketing stunt. Anyone heard of DNSSEC mass deployments so far? The world ain't ever gonna be immaculately perfect. Hear hear.
Nothing to see here, move along.
Just for setting this clear: ;)
The biggest security hole to the internet is the human beeing!
A man-in-the-middle attack on BGP would require that you intercept and re-write BGP data. The only place to do that is if you can insert some hardware on the physical route between two BGP-speaking routers. That is, on the cable between two ISPs that are peering with each other or have a transit agreement. While the BGP protocol could, in theory, be routed across the internet, my understanding is that in practice it never is.
Add to that that to successfully perform such an attack, you would need appropriate (expensive) network interfaces and hardware capable of speaking fast enough, and this "attack" becomes something that needs a *lot* of resources to pull off. Sure, governments and big corporations can do it, maybe big organised crime could too, but yer average bedroom cracker couldn't.
And why would the big boys bother anyway, when they can just announce bogus routes?
BGP poisoning enables a man-in-the-middle attack; encryption will only work with either a pre-shared key, pre-shared identities, or a trust infrastructure such as ssl.
Unfortunately most protocols don't have these as options; if they support encryption at all, it's with neither any explicit prior knowledge of the other party nor trusted certificate authority. This means that passive sniffing is blocked, but man-in-the-middle is just a little harder than it would have been, but still entirely feasible.
Slay a dragon... over lunch!
And you told the government how to do it? The government that has changed FBI regs to allow spying on citizens not suspected of crimes? The government involved in that whole illegal wiretapping?
Thanks...
Asshole...
The only change I can believe in is what I find in my couch cushions.
You could claim to have a better path from Bank of America's SMTP servers to Time Warner Cable's SMTP servers. Then, request password resets for a bunch of accounts and intercept the emails with the tokens.
As long as you can handle the bandwidth for that small slice, you've done it.
It's not universal, but a substantial number of ISPs do filter the routes of their customers. A few of us gave a tutorial on the subject at the NANOG 43. Unfortunately, most ISPs don't filter their peers, so once a bad route gets injected somewhere, it'll tend to get passed on repeatedly. The IRR projects, as well as both SO-BGP and S-BGP are designed to try to mitigate that.
Need Geek Rock? Try The Franchise!
The NSA doesn't need stupid BGP tricks to eavesdrop on Internet traffic. They can take over rooms at peering point and put in optical splitters to their heart's content.
Do not fold, spindle or mutilate.
you're probably right. They probably put the BGP bug in there in the first place.
They're using their grammar skills there.
Can anyone who is familiar with this tell me whether this is an IPv4 specific issue, or whether it would impact IPv6 as well?
Jumpstart the tartan drive.
"'I went around screaming my head about this about ten or twelve years ago... We described this to intelligence agencies and to the National Security Council, in detail.'"
That's some smooth work right there. Later that year he also explained in detail to a guy in a mask and trenchcoat that the vault at his local bank had a serious flaw, allowing access to anyone with a little time on their hands. I'm sure that guy got right on it.
HitScan
Secure communication over untrusted medium is a solved problem.
Hell, Diffie-Hellman key exchange is older than I am.
So, the moral is that any protocol that cares to protect itself from MitM
attacks can. The current problem is that many protocols currently in use
don't bother (either from laziness or from the fact that they were
designed during a more trusting age).
*sigh* back to work...
This exploit doesn't alter the behavior of BGP in any way. What this exploits is that there is no existing capability in the BGP protocol to sanity check a prefix and its announcer against an authoritative list of prefixes and owners on the fly. This can however be accomplished manually via prefix filters on BGP peerings.
http://www.cisco.com/en/US/docs/ios/12_1/iproute/command/reference/1rdbgp.html#wp1175412
Some attempts have been made to create authoritative route registries from which filters could be updated via scripts, but they are not necessarily used everywhere. Note the email discussion in the faq below is eight years old.
http://www.irr.net/
http://www.irr.net/docs/faq.html
In conclusion, handwaving and drama queening aside, this is nothing new and solutions to the problem exist.
If hes legit, please do. Before lunch if you don't mind.
---- Booth was a patriot ----
I think it will provide a good way to get good marketing data. For instance count the number of Windows Updates by Windows version to see what the true market share of 98/2000/XP/Vista really is. See the market share of various Linux's and other programs that automatically update or call home. Would be good to see real data, since these would be systems in use and not shelfware. Much better than relying on estimates by places like Gartner or Forrester. I'll have to get the market research department on this right away!
If you don't encrypt.
---- Booth was a patriot ----
Diffie-Hellman is vulnerable to man-in-the-middle attacks if you don't have a way to verify the signing keys of either side (that's what I was talking about in my previous post).
It provides encryption, not endpoint authentication.
Slay a dragon... over lunch!
Sorry, no! You are working under the same premise of the routers. You are thinking as a BGP router does. In reality you only have to lie that you have a better path and you only have to lie that you have better bandwidth.
Basically, they're just feeding gullible BGP routers with /32 host routes.
It may indeed be a "blunt instrument", as you say. But, hammers are blunt instruments and they, like this exploit, work exceptionally well.
These guys have demonstrated and publicly proven the technique to work effectively. Your lack of "vision" doesn't make the vulnerability any less real or the exploit any less effective.
Maybe he did it as a patriot. If this was discovered by other countries first they could use it against us and we wouldn't have a clue about what was going on or where to start fixing it.
There are numerous IP networks out there. Some of them eg: SIPRNET referenced above, are private networks. There are methods to insure the privacy and integrity of these elements and devices. The same can be done and said about some corporate security practices. The drug companies are very careful about protecting their secrets so they can obtain patent protection. The same general model applies to classified communications as well. You have protection for a period of time, then after that it expires. Think of all the documents that have reached their "declassify" date in the past few years, from the Mob+Castro plot to others.
But yes, there is more than one network built with the Internet Protocol (IP) networking technology.
In a bid to play 'Mr. Network Engineer Smarty Pants', it seems you failed to find the time to read the actual presentation slides. It doesn't claim to alter the behavior of BGP. It simply exploits the lack of security-conscious network engineers and/or registration security holes.
"The purpose of learning is growth, and our minds, unlike our bodies, can continue growing as long as we live." - M.J. A
I've worried about the possibility of BGP poisoning attacks myself, but only because we have a virtual monoculture ...
BGP is a near-monoculture because the standards document is not adequate for buidling a working implementation, while a broken one breaks not just the box running it but much of the net (and essentially all of it that has traffic that should go to or through that box).
The protocol has negative feedback and delay, which results in oscillation ("route flapping") if nothing is done about it. The current workarounds for this are arcane and not well known.
Because deploying a broken BGP implementation breaks things so badly, ISPs are very loath to switch to boxes that don't have a proven BGP implementation. So one of the biggest barriers to entry for an equipment manufacturer is licensing customer-trusted BGP code or finding the rare talent with the occult knowledge, building their own, and then convincing the ISPs that the implementation is safe.
As a result there are VERY few BGP implementations deployed. Last time I looked - a few years back - there were three: The original (don't recall who did that one), Cisco's, and Redback's. Getting its BGP accepted was a significant part of why Redback's Smartedge product line was slow in breaking into the market.
This near monoculture also means that, in addition to the exploits inherent in the protocol, any implementation-specific security flaws will have a very large target population for exploits.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
I find it very interesting that in this day and age, some "security researcher" can present on this subject and illicit an OMG-end-of-the-world reaction from people (primarily the press).
BGP was designed to work this way. The "hack" they described is not a hack but a key feature of the system. There is an implied level of trust one must have with their peers. When that trust breaks down, or isn't there to begin with, trouble can be found.
Yes, there are methods that can be put in place to prevent this, e.g. filtering at the end of the provider networks. There are cases, however, where that may not be option. Also, it is a point that while one nonconformist to proper edge filtering is bad, it is certainly better than everyone not filtering.
Other things put in place to help the filtering include the Routing Arbiter Database (radb.net). This allows providers and their peers to publicly define the address ranges that will be expected to originate from a given AS. Automation can be performed against these datasets, and filters generated on the fly. Many providers do this already.
Just because not everyone does this, does not mean it is the end of the world. If someone does exploit this, it is usually pretty easy to see where it is coming from. I've seen it happen before myself.
Nothing new here, guys... move along.
What about if someone like MediaSentry, or whatever name the MPAA/RIAA jocks have this month, used a BGP attack to redirect all of your download traffic via their Internet connection so they could spy on all of the tunes you're downloading?