BitTorrent Devs Introduce Comcast-Proof Encryption

Dean Garfield writes "An article at TorrentFreak notes that several BitTorrent developers have proposed a new protocol extension with the ability to bypass the BitTorrent interfering techniques used by Comcast and other ISPs. 'This new form of encryption will be implemented in BitTorrent clients including uTorrent, so Comcast subscribers are free to share again. The goal of this new type of encryption (or obfuscation) is to prevent ISPs from blocking or disrupting BitTorrent traffic connections that span between the receiver of a tracker response and any peer IP-port appearing in that tracker response, according to the proposal.'"

Do arms races ever work?

[pembo13 (770295)]

Unless one side suddenly blows away the other, I don't see this ending. It may breed innovation, but said innovation only seems useful for this one problem.

Re:Do arms races ever work?

[webmaster404 (1148909)]

Well, its not an "end-all" solution however it solves the immediate problem. However chances are in 10-15 years we won't even be using Torrents we will have moved on to another form of P2P.

Re:Do arms races ever work?

[by Anonymous Coward]

Why wait 10-15 years? Jump on the bandwagon and make impossible predictions about the near future.

In 10-15 years, p2p will stand for Person to Person, as we will have placed the computers inside our heads, we will share thoughts. No more picture based porn, when you "download" the new porn, it will appear as you in it. And you will not only get to see/heard, but also smell, taste, and feel. More importantly, cyber-sex will be much more like real sex, as a virtual world will be just as real as the real world.

Oh, and in 20 years legislation will have been past severely restricting this new technology to anyone under 21 years of age, and in some states, cyber-anal-sex will be a capital offense. In 23 years, Comcast will start 'degrading' this new service for due to 'QoS' concerns. After a few million people have their virtual parters turn into cows during virtual sex, a riot breaks out leaving America as a second world nation.

Re:Do arms races ever work?

[fyrewulff (702920)]

Yes. Once they actually do make disc based media that can actually take a fall, we'll be using the FDTP (Flying Disc Transfer Protocol) method.

However, the packet drop in windy places would be too much.

Re:Do arms races ever work?

[linzeal (197905)]

We are still using HTTP and FTP, who is to say that BT will not just slowly mature like those? If there is any standard P2P protocol emerging than BT would be in the top 3 along with Edonkey and DC++.

Re:

[Ilgaz (86384)]

Unless one side suddenly blows away the other, I don't see this ending. It may breed innovation, but said innovation only seems useful for this one problem.

As far as I followed, most Bittorrent based "inventions" were done because of attacks by dark companies (media defender), fake seeders etc. Comcast is practically DOS attacking their own customers so someone finds a workaround for it. If it is good enough, all those bittorrent clients will adopt it in no time and they will end up with horrible publicity, paranoid customers, FCC investigation for nothing. Technical karma :)

Re:Do arms races ever work?

[moderatorrater (1095745)]

Do arms races ever work?

Depends on your objective. Generally, arms races preserve the status quo, which, in this instance, is exactly what they're trying to do.

Re:Do arms races ever work?

[azgard (461476)]

Depends on your objective. Generally, arms races preserve the status quo, which, in this instance, is exactly what they're trying to do.

The question is, what is the status quo? Is it the filtered or the unfiltered internet?

Re:Do arms races ever work?

[rale, the (659351)]

Comcast's bittorrent filtering has almost certainly cost them money in the form of hardware and software to implement it. If continual updates to the protocol make it more difficult and expensive to filter, then theres always the chance that ISPs could decide it's actually a better investment in the long run to upgrade their networks, rather than upgrade their filtering. That could just be wishful thinking, tho...

Re:Do arms races ever work?

[MightyMartian (840721)]

Or they could just do the sensible thing, cut out all the bullshit "unlimited" advertising (which should be against the law anyways) and start selling customers a set block of gigabytes, with an over-limit charge per gig, just like the dialup ISPs did with time online in the olden days. That's what I did at the small ISP I worked for. I wrote and maintained the billing software, and just sucked in usage stats off our Radius servers once an hour. The system was even set up to send out an email when a user was close to his gigabyte limit letting him know that the meter was going to start running and what the charge per gig was.

We tried shaping P2P traffic, and it just annoyed customers, and annoying customers is not exactly a long-term strategy for success.

That would be suicide...

[Joce640k (829181)]

If they ever do manage to completely block P2P then they might find themselves looking at a bunch of customers who only want 300kbit connections instead of 20mbits. What are they going to do? Slash their prices to the same as the small ISPs who can offer cheaper/slower connections? I think not.

Re:Do arms races ever work?

[Zebra_X (13249)]

"Or they could just do the sensible thing, cut out all the bullshit "unlimited" advertising and start selling customers a set block of gigabytes."

I can assure you, you don't want this. You assume that the ISP's are going to give you a "reasonable" block of data to transfer on a monthly basis and a reasonable price - they are not. They will use this pricing scheme to "extract value" from their customer base in the form of quotas that are properly tiered so as to be just below the common usage tier. The result will be many customers need to go a step higher, and are charged more, for considerably less than they had access to before. Do you really want to worry about whether the next movie you get off of iTunes is going to pop your quota? Or the next stream you setup?

Honestly, bandwidth in the US is what is causing a great deal of innovation at the moment - look at iTunes and Netflix now offering entire movies as either downloads or streaming. Caps will only stifle the adoption and innvoation of this type of technolgy. Customers will think twice about the double cost of streaming a video - the cost to their cap, and the cost of the service. There are I'm sure other bandwith based applications out there that we have not even thought of.

The answer is just in disclaiming that running certain types of services like bittorrent coupled with excessive transfer on a connection can lead to service degredation, not termination. They just need to put a process in place to handle this situation. Time warner claims that "5% of their customers use 50% of their bandwidth" - well - that seems pretty damn easy to fix doesn't it? Exceed a certain monthly transfer rate, send out a warning via e-mail - usage continues - put a cap that is far lower than their original amount.

In addition they don't really say that they are running out of bandwidth, so I'm not sure I see where the problem is.

Re:Do arms races ever work?

[ScrewMaster (602015)]

In addition they don't really say that they are running out of bandwidth, so I'm not sure I see where the problem is.

They're not ... they're running out of shareholder satisfaction. Their customers are demanding more capacity, and their shareholders are demanding more money now. The two are diametrically opposed, with the ISP squarely in the middle. Either we adjust our expectations downward, or the shareholders do.

Who is the most like to get what he wants?

Source of the unsolicited traffic

[Lonewolf666 (259450)]

Reportedly most of it comes from botnets:
Insecure machines that were taken over by hackers and whose clueless owners did not notice anything. Or even don't care.

Now if ISPs start selling traffic by the gigabyte (again - it was not uncommon a few years ago), the owner of those spam-slaves would notice it on their internet bills. At that point, I think securing one's machine would become a lot more popular and the botnets would shrink. Overall result:
less spam and DDOS attacks.

Considering the inbound hacking attempts, my father still has a 2 GByte/month plan and so far I've heard no complaints about suddenly increasing bills. So it seems to be not that much.

Re:Do arms races ever work?

[irc.goatse.cx troll (593289)]

Define 'connection'.

All you would need to do to circumvent that is use something stateless like UDP. If they want to limit UDP to something like no more than 100 different IP's sending you packets within a set time period, they just created an amazingly simple DoS attack against all of their customers.

Even without udp you could just make sure you fully close all your connections as soon as possible, if not sooner (i.e kill slow clients to make room for fast ones).

Also setting this too low could limit legit use, like when you start up your computer and have a burst of all your software checking for updates, checking for mail, rss feeds/podcasts/etc going off, all your IM clients connecting to their various servers, etc.

Re:Do arms races ever work?

[CodeBuster (516420)]

The bittorent devs have the upper hand, at least for the forseable future, because of strong crypto like AES, Serpent, and Twofish for symmetric session traffic and strong public key crypto like RSA to handle the handshakes and symmetric key exchanges. The only response of the ISP is to try and automate Man in the Middle (MITM), but that will be extremely difficult and expensive to implement in practice. Remember that Comcast was throttling bandwidth to cut costs on network upgrades so why would they spend exponentially more on new specialized crypto hardware and software to MITM the handshakes on bittorent sessions if they are too cheap to even upgrade their network? Unless and until there are substantial advances in cryptanalyis (as far as I know there have been no substantial improvements on known attacks in recent years, minor optimizations here and there but not enough to really put a dent in the crypto) or quantum computers become cheap and practical, encryption will provide a very strong defense against network filtering, particularly when it is combined with port randomization. That is why it is in the best Interests of Comcast and other ISPs NOT to escalate by engaging in packet filtering. They will only hasten the development of bittorent clients with strong crypto, as they are doing here, AND draw attention to these new "super" clients that are not "slow".

What does strong crypto have to do with it?

[Joce640k (829181)]

Anything stronger than rot-13 will do.

Even if it only takes an ISP 0.1 seconds to "crack" a packet then there's no way he can crack the millions of packets per second flowing through his routers.

Re:Do arms races ever work?

[Just Some Guy (3352)]

Remember that Comcast was throttling bandwidth to cut costs on network upgrades so why would they spend exponentially more on new specialized crypto hardware and software to MITM the handshakes on bittorent sessions if they are too cheap to even upgrade their network?

That's a very important point. Comcast is going to have to spend $X to make their network tolerable, either by buying blocking P2P and other bandwidth-hungry application, or by expanding capacity. The first method gets them a nice, controlled, slow network and the hatred of all their potential customers. The second gives them a wild-and-woolly, fast network their customers love (and therefore more customers). So, again, given $X: do you invest it to lose business or gain business? That's really the choice here.

Given Comcast, they'll probably use it to put ultrasonic speakers on their modems so that teens don't want to use them, then five years lateer ask Congress for a bailout because they're uncompetitive.

Re:Do arms races ever work?

[madsenj37 (612413)]

1. Evolution is an arms race. Viruses and bacteria attack us and we adapt, so they adapt, creating a cycle.

2. Free markets are an arms race. When one business evolves, the other must to survive or perish.

Traffic Analysis

[gaika (975356)]

Most blocking systems use traffic analysis to block encrypted protocols, even the ones pretending to be something else. There's no way you can confuse p2p sharing with normal browsing if you look at the pattern of data flows.

Re:Traffic Analysis

[Azh Nazg (826118)]

That's nice, except that blocking encrypted protocols blocks quite a bit more than BitTorrent. . . Secure banking over SSL, SSH, VPNs, and a whole plethora of other protocols. Unless an ISP is willing to go from Internet Service Provider to Web Browsing Service Provider, it would be foolish to block encrypted protocols.

Re:Traffic Analysis

[budgenator (254554)]

that's what the cableco's really want, they can easily oversubscribe the system when all you can do is browse the web and Email.

Re:Traffic Analysis

[gaika (975356)]

Nobody is going to block all encrypted protocols, that's stupid. They identify the application that is using encryption by looking at the shape of the traffic flows. p2p apps open tons of connections, exchange about equal amount of data both ways, and have a distinct negotiation phase.

Re:Traffic Analysis

[Not_Wiggins (686627)]

I think you may have missed the point of the GP post.
The point wasn't to block encrypted traffic just because it is encrypted. It would be to do traffic shaping, so that a connection generating dozens or hundreds of simultaneous encrypted connections to different destination IP's might be targeted; it is a traffic pattern would most likely be generated by a P2P program and not by normal internet use by a family.

Re:Traffic Analysis

[jonwil (467024)]

Banks dont use consumer grade internet connections to talk to each other.

Re:Traffic Analysis

[ookabooka (731013)]

I think the idea here is to stop Comcast from injecting their own RST packets into the stream, effectively killing the connection from both sides. Every time an ISP implements a harsh countermeasure, they force the evolution of the protocol. I see this simple as the next logical step in the constant pull and tug of P2P and ISP's. Still, kudos for these guys doing this stuff. I'm sure Blizzard will like hearing that their updates are hindered on Comcast's networks while P2P data has an easier time.

Re:Traffic Analysis

[by Anonymous Coward]

Actually, IPSec will prevent the ISP from being able to reset the flow. If a packet comes in that is not signed/encrypted (depending on the mode) with the credentials of the other end-point, it is discarded as an attack. It's a pain to set up IPSec security associations in many conditions, but IKEv2 has made it somewhat better.

The fact that you are buying service from the attacker doesn't make them not an attacker. The counter measures developed to fight attackers may have limits, but they are there and are useful in this context.

Another volley herd in The Pirate Bay

[corsec67 (627446)]

Too bad we even have to fight this forgery by Comcast, but a technical option has its advantages, since a legislative option might get watered down by lobbyists and congress.

Encryption is always a good thing. The more people that use encryption, the less eavesdropping there will be.

How about, "if you have nothing to hide, hide it anyways"?

Re:Another volley herd in The Pirate Bay

[webmaster404 (1148909)]

How about, "if you have nothing to hide, hide it anyways"?

How about, if you have nothing to hide, someone either the government, your boss, Etc. will twist it to either sell your info or make you look like a criminal, so hide it.

Re:Another volley herd in The Pirate Bay

[Sir_Lewk (967686)]

How about, "Since I have nothing to hide, you shouldn't mind not reading it"

Re:Another volley herd in The Pirate Bay

[mdmkolbe (944892)]

If I have nothing to hide, you have no good reason to read it.

doesn't work

[nguy (1207026)]

Comcast will now probably simply impose soft traffic caps and soft caps on the number of connections users can make.

Re:doesn't work

[Wildclaw (15718)]

Actually they are doing it because they have an outdated badly scaling last mile network and don't want to spend the nescessary capital to improve it.

There is a reason that it only is cable companies talking about bandwidth caps, and not the dsl companies.

Comcast makes $$$$$ disrupting seeds

[colinmcnamara (1152427)]

Comcast is trying to spin their actions as promoting fair use of the their networks. The truth is that ISP's profit from having data dumped INTO their network and have to pay hard cash for data LEAVING their network. By injecting RST's into the peers seeding traffic, they promote an asymmetric data flow that brings more data (and therefore money) into their network, while minimizing the money they have to pay other ISP's for data going out. This proposal provides protection against the throttling of their upstream Bittorrent traffic only if the ISP is not aware of the info_hash of the torrent. Once this data is known it is possible to apply common data tagging and congestion control techniques to squelch this traffic. All the service provider (or application developers like SandVine) has to do is monitor the common torrent sites, and dynamically update this hashes into the network filters. This is sure to deny a majority of the torrent traffic out there (movies, linux distro's, etc). Colin McNamara CCIE #18233

Holy crap, a CCIE!

[by Anonymous Coward]

I am just a measly CCNA.

I am not worthy.

m(_ _)m

Re:Holy crap, a CCIE!

[by Anonymous Coward]

I for one find anyone flaunting certification X to be an annoying twat

Re:Holy crap, a CCIE!

[caluml (551744)]

I for one find anyone flaunting certification X to be an annoying twat

I agree with that.

Professor Sir Calum, MP, PhD, MsC, Esq.

Ha! Ha!

[stox (131684)]

Now Comacast will need to keep a list of connections in order to guess that a torrent is running, instead of just looking at the packet. Good luck on that without a massive infrastructure upgrade.

I'm glad this is all happening

[bogie (31020)]

It had to come to a head at some point. ISPs have been bitching about P2P for a while now. Let's get those secret docs on "unlimited" usage out in the open. Let's define what is acceptable and let's give users the ability to meter their usage. My prediction is 95-99% of us won't be affected by these new open bandwidth policies and ISPs can go back into the business of providing dumb pipes.

Throttling encryption

[diamondmagic (877411)]

How long is it until they start throttling encrypted traffic too?

Won't work: They clamp on traffic per flow

[ZWithaPGGB (608529)]

They don't care about any protocol analysis. Any sufficiently long-lived, high volume, traffic flow between two IP addresses gets hit. I've had IPSEC VPN connections behave strangely and opened tickets, where the techs have admitted I had "accidentally" been flagged (IE, the IPSEC endpoints weren't on the whitelist, even though I have business class service).

The only way around this is to open multiple connections to different addresses, transfer small amounts per connection, and then shut it down, opening the next connection to a different endpoint. It requires a total reengineering of P2P, although the BitTorrent mechanism is closest to what would work.

Ultimatly it wont stop comcast

[jonwil (467024)]

If they aren't already doing it (I dont know the exact technical details of what they are doing), ISPs like Comcast will simply start looking for anyone uploading large amounts of data (especially if they are uploading to a bunch of different people at once) and block that.

First Blood?

[EdIII (1114411)]

I'm surprised it took this long for the Bittorrent Devs to respond. Encryption is not a complete solution, as I have stated before, but it is a beginning. That is for certain .

It's going to get a lot more interesting from here on out. In the end, it will only benefit the consumers since they will receive technology that allows them to communicate a little more privately, and perhaps with a little luck, more anonymously too. One could only hope that TOR/Freenet technologies become as ubiquitous in their use as email. Perhaps a hybrid system with elements of Freenet, TOR, and Bittorrent all wrapped up into one would do the trick. I certainly think so.

I think, actually I know, that Comcast has fired the first shot in a losing battle.

I also just can't help pointing out the similarities to the Drug War. A million or so people in prison, and yet there are still plenty of users and suppliers. I would almost say it has effectively made no difference in the amount of people using drugs, or selling them. Especially, since the amount of drugs being sold and used in prisons is even higher then on the street.

So what is the point? If history has taught us anything, it is that governments (corporations even more so) will consistently fail at their attempts to limit/eliminate popular behavior. The elements may change from time to time, but the end result is always the same. The people will find a way to continue their behavior .

"Greetings, Professor Falken. Strange game. The only winning move is not to play."

Re:First Blood?

[dave562 (969951)]

The point is to generate revenue by exploiting people's natural tendencies. Think of all the fines to be collected. The reconnection fees. The court fees. The jobs generated tracking torrent users. The training programs to be created to teach the fascists what they are looking for. Just like with the war on drugs, the point isn't to fix the problem. The point is to so fully integrate the "problem" into the system that it serves as a source of energy for and an excuse for the continued existence of the system itself.

Re:First Blood?

[EdIII (1114411)]

Eating up crazy resources beyond the convections of a normal average daily internet regiment

ISP's are inundated with mass amounts of data chocking off more important services

I'm sorry, but you are dead wrong on that one. 100% Absolutely, Positively, Infinitely WRONG. I hear a lot of people say that. I cannot possibly be wrong in my usage of my connection for the following reasons:

1) I pay for it.
2) It is unlimited.

unlimited (n-lm-td) adj. 1. Having no restrictions or controls: an unlimited travel ticket. 2. Having or seeming to have no boundaries; infinite: an unlimited horizon. 3. Without qualification or exception; absolute: unlimited self-confidence.

They set those terms, not me. They have continually advertised a position that was in fact the opposite of their true intentions. All that matters is the contract here though, and that states unlimited.

I don't know if English is your first language, since your use of grammar is a little off, which I don't say in a negative way at all. I just don't understand what you mean by "convection".

You say "normal average daily internet regiment". That is in of itself, an observation only. It is meaningless to the discussion since it just a statistic. No one is actually bound by contract, or any verbal representations by any ISP that they must maintain a normal level of use. Unlimited means that you cannot apply any limitations on the usage; "Normal" is a limitation.

You also talk about more important services. There are no "more important services". Everybody is unlimited, therefore all traffic is equally unlimited. The ISP must therefore treat all traffic the same according to the representations of an unlimited contract.

Now if at some point in the future, the ISP offers for people to voluntarily apply QOS principles to their network traffic, that is in the best interests for everyone. I have no problem being asked, nicely, to apply a QOS tag to all my communications, as it only helps me in the end. I also like the idea of being nice and cooperating with my neighbor, so that under heavy load conditions, his VOIP sessions will get the priority he needs. The contracts could redone to reflect this in the future.

What about the collateral damage?

[blake182 (619410)]

One of the things I'm curious about is what kind of collateral damage this kind of thing does to legitimate traffic. Oddly enough, I couldn't get to expedia.com, transformers.com (hey, I have an eight-year-old), and store.apple.com when I first got Comcast. A couple of months later, when the news first broke that they were screwing with the traffic, those sites suddenly started working. Nothing changed at my house, and all of them started working at once.

Possibly coincidence. Possibly not.

Re:I wonder...

[Kadin2048 (468275)]

Well currently the state of the art is in favor of encryption, rather than cryptanalysis, so I don't think that the advantage is automatically Comcast's. They could probably do some fairly sophisticated traffic analysis, but at the end of the day, they're not actually going to break the encryption and get at the contents, and they can't block all encrypted traffic because it's too critical for other purposes.

They can force the BitTorrent devs to produce a new version every few months, but in the long run I think they're on the losing end of the war -- if they want to stay in the data-transportation business, and assuming there aren't any major breakthroughs in cryptanalysis that render modern public-key technologies useless.

Re:I wonder...

[MrShaggy (683273)]

Wouldn't breaking the encryption be a violation of the dmca?? I think that would be sweet.

Re:I wonder...

[budgenator (254554)]

there is also a UDP Tracker Protocol for BitTorrent [bittorrent.org], UDP doesn't even hear the RST packet. Comcast will have to figure out a way to turn off something that doesn't have an off switch.

Re:I wonder...

[Mr2001 (90979)]

Nope. It's the TCP connection between two peers that Comcast is attacking, not the connection between the peer and the tracker. Using UDP for the latter doesn't solve anything.