Oyster card security broken.

Oyster cards are cards containing RFIDs which facilitate transport on the London Underground (and overground) networks. I had been wondering for a while how exactly they worked, but not finding much detailed information online I based my conclusions initially on my observations.

Here is what I have deduced.

1. Each RFID card has a unique ID which is recorded during each transaction with a card scanner.

Any Oyster user can access their usage history, either online or via a Oyster top up machine. This presents an interesting problem - if you can check up on where your card has been, what is to stop Transport for London from using the same information? Either individually or as part of the collective, it presents a very detailed picture of individual and mass use of the transport network.

2. The RFID works passively and contains a small amount of EEPROM.

Given that the Oyster card needs to be ‘tapped’ on the reader every time, it is safe to assume that the RFID does not have an internal power source. Instead, it only becomes ‘active’ with the energy it obtains via induction from the electromagnetic field close to the surface of the Oyster touch point. This energy is sufficient to power up the (presumably) CMOS device which then sends the encrypted data to the reader. It is not clear at this point whether the reader then sends back a response with the new balance to the card, or whether the entire ‘transaction’ process is done on the RFID card.

3. The information stored on the EEPROM is encrypted, most likely with symmetrical cryptography.

4. When scanned, the information from the Oyser card is used, it is not pulled from a central server.

When updating the Oyster card the card itself must be touched against a scanner. If this is not done the balance is not applied. I initially believed all balance and travel card information was securely stored on a ‘mothership’ server. This clearly can’t be entirely the case. Although, when a top-up is bought online, it is stored in the Oyster system until the Oyster card is touched on a reader somewhere in London. This suggests there is a ‘mothership’ server which records all this information, although it is likely it is only linked to newsagent kiosks and top-up points, not the barriers themselves otherwise there would be no need to store the information on the card.

5. Not only can the RFID store a balance, it can also store season tickets for a variety of durations and zone validities.

However, the title of the post suggests the security is broken, and indeed it is, although not through my investigations. A Dutch team took this a step further.

It turns out almost all my assumptions were correct, the Dutch team used a portable device to ‘touch-in’ on an Oyster reader, this disclosed the encryption key used on the Oyster device which they then stole. In possession of this, not only could they decrypt any Oyster card to determine how the information was stored but they could also theoretically generate any balance or season ticket, which encrypted properly would be indistinguishable from the real (paid for) thing.

However, to avoid no doubt countless hours of reverse engineering, the Dutch team brushed up against commuters on the tube and wirelessly interrogated their cards, stealing the information that was on them. This allowed the team to effectively clone cards which were valid, entitling them to free travel.

But the story does not end there, it turns out the company that makes the RFIDs for Oyster cards is called MIFARE, and their chips are used in a wide variety of sensitive installations in a variety of countries.

When the file extension… is not the file extension.

I was bemused to read on bbc news earlier that a trivially simply ploy stung half a million file sharers. The concept is nothing new having been started a fair few years ago by virus / malware writers and adopted by Copyright enforcement agencies in recent years. Do the anatomy of a decentralised file sharing system, anyone can seed a file. Once this seeded file is made available to the peer-to-peer network it either becomes advertised to a localised central file distributor (referred to as a Super Node or Server) or is found during a spider search query run by another user logged into the peer to peer network. If these files are topical or sought after, they can be transferred onto a different node (client) rapidly. There they are stored in the second user’s ’shared’ directory where more people can download it.

Once a seeded file has been downloaded and spread over a few tens of nodes the rate at which it can be downloaded by others increases almost exponentially with a cascade like effect. Other people of the peer to peer network are lured into downloading this file based on the number of people who have it therefore assuming it must be genuine and would be comparatively quick to obtain. Couple this with a topical or sought-after song / album or file aimed at the masses (who statistically would contain a fair percentage of PC-illiterate users and those with a penchance for agreeing to all the pop ups they come across) means these files explode across networks.

This malicious file in question appears to have masqueraded as a MP3 by Girls Aloud. Given the fact that on running the file pops up a message saying the computer requires a codec to play the song and tries to direct you to a website in order to download it, most computer users would stop and reexamine what they had just downloaded. People that brazenly proceeded and downloaded the malicious ‘codec’ package had spyware installed on their system which would ‘bombard’ users with pop ups. Also, the download file would spawn copies of itself within the User’s shared folder under different names to try to make itself attractive to a greater audience.

But what happened? How were people tricked into downloading an MP3 file but ended up running a malicuous program? The answer to this lies in the file type. Broadly speaking, there are two ways in which a file can be opened:

1) via script or binary execution (e.g. .exe, .com, .vbs, .java, .scr … and some others)

2) via program read from an external application (e.g. .txt, .doc, .wav, .mpg, .avi …. and MANY more.)

MP3 files (Moving Picture Experts Group version 1 audio layer 3) are the latter, upon execution, Windows searches through its list of known file extensions stored in the registry to see what it should do. It instantly finds the entry for MP3 and sees this type of file is handled by a media player like Windows Media Player, WinAMP, iTunes etc etc. Windows then executes the media player which, on loading, opens the MP3 file specified in the command line argument, decodes a block, fills its buffer and starts to play. Unless a clever trick like a buffer overflow is used, which have historically been responsible for security breaches in various Windows programs as well as console homebrew development, this renders all ‘program read’ type files harmless*. As such we have to look elsewhere for the source of this problem.

That brings us nicely to the point I wanted to raise in this post, file extensions and more specifically, security vulnerabilities in their implementation. Recent versions of Windows from XP (and possibly earlier, I can not remember) have automatically hidden the file extension by default leaving the user to distinguish between file types by iconographic representations. Whilst at times this is both cleaner looking and more functional, it does present an interesting security problem, what if there are two file extensions? Window will quite happily truncate the file .xxx from a file name leaving the first extension, despite the fact Windows ignores anything before the final .xxx . As a result, if you name a file SomethingInteresting.mp3.exe, in its default state, Windows will happily display the file as SomethingInteresting.mp3 but will execute the file as an EXE when double clicked. Obviously, if you quieried the file by right clicking on it and selecting properties you would be immediately told what type of file it is, but most people will take the file at face value.

Luckily there is a very simple way to gaurd against such black magic, in Windows XP and Vista** in the file browser, goto the Tools menu and select Folder Options.

In this dialog, uncheck ‘Hide extensions for known file types’ and click Apply followed by clicking Apply to all folders.

And that’s it! A simple check box and some common sense now separates you from being lured into downloading fake or malicious files.

* Some files like some movies can have containers which direct the media player or operating system to web pages. It is not just media files which are vulnerable but this is a completely different topic.

** In Vista you may have to enable the classic menu

Test your password!

I found a fantastic site today which quickly and transparently tests the strength of your password(s). I would not recommend you actually type your current passwords in (just in case there is some sneaky javascript code) but it is interesting to experiment with different password combinations.

Since I was on the topic of passwords, I ended up writing a brief post about how to choose a good password and general password security.

A good password should be four things:

1) Use at least two cases* (e.g. lower case, upper case, ‘number’ case and ‘character’ case.)

2) Be a suitable length - anything less than 7 characters should be avoided.

3) Not include repetition within the password and should not be used for more than one application.

4) Be something personal or easy to guess (a birthday, pet or family member name or related to the application - for example ‘email’ as a password for your email account would be ludicrous.)

Let look at some examples:

The old favourite: “password”. As you can see from the rating below, it is a terrible password. Not only is it predictable (and one of the most commonly used passwords) but it uses only one case and has some repetition (sequential double ’s’.)

A slightly better version of the old classic: “pa55word”. This time, all I had done is replaced the ’s’ with the 733t-ified version. By adding numbers, the complexity of the password has increased dramatically although it is still hindered by repetition.

Lets go even further: “Pa55Word”. Now we are using three cases and the result is predictably much stronger than using two cases alone.

And finally, lets go nuts: “Pa5!Word”. Using all the cases available on the Roman alphabet and removing all sequential characters. It is still not a brilliant password, but it is head and shoulders above the others.

Whilst choice and selection of password is important, it is not always essential to pick random strings as your password. Whilst passwords like gY$5c0p[ are very strong (it scored 92%) it is difficult for most people to remember them due to their entropic nature. It is therefore important to marry practicality with security and my advice to anyone picking a password would be to think of a word (or phrase) and substitute some of the letters for numbers / capitals / characters as in the example above**.

Password hints:

1) If you are choosing a very important password, pick a passage from a book. For example, the first 3 (or as many as you want) words from the first line of a particular page** and add a good degree of randomness to it as described above. If you need to jog your memory in the future, simply refer to that page and it should normally come back to you.

2) If you must write or record your password, obfuscate (via a stenographic method) it! Split it in half (or more pieces) and hide the password/passphrase in several bits of innocuous data. For example: If you made your password Nice225 Woods987 then you could store the following contacts somewhere:

William Nice +44207 750 1225

Christian Woods +43133 987 3245

The same method can be applied for card PIN numbers which can be stored as part of a dummy contact on a mobile phone.

3) Never stick to the same password for more than one service - if someone compromises one password, all your services will be vulnerable.

4) Scale your password to the particular security environment. A password that is used for an unencrypted email account need not be as strong as one for a SSH / VPN / Remote Terminal or VNC account.

5) For accounts you are particularly cautions with, rotate your password frequently. This need not be very week or even every month. If you change your password every 2 or 3 months, it will provide a much better protection against online stalkers who may be lurking and checking your accounts / emails periodically.

6) Passwords can be passphrases! It is much easier to remember a line of a story / poem etc than a bunch of rubbish. Unfortunately, even if that line of text is long enough, it will not offset the problems** caused by character repetition, although it would be important to obfuscate it in some way.

Appendix

* The reason cases are so important is simply a matter of maths. If an attacker knows the password is only one (or two) cases, it significantly reduces the amount of computational time to brute force (or guess) the password. Take for example, a password with only one case (lets assume its lower case). There are only, 26 characters in the Western (Roman) alphabet meaning the complexity of the password is:

…if the password is 4 characters long, there are : 456976 combinations.

If the password is 8 characters long, there are : 208827064576 combinations.

Now lets assume two cases (lower and upper case) are used. Now the attacker has to try a total of 52 character combinations for every character suspected to be in the password.

…if the password is 4 characters long, there are : 7311616 combinations.

If the password is 8 characters long, there are : 53459728531456 combinations.

You can quickly see the significance in the numbers. If to round it off, we try all the (printable) characters available (94), an 8 character long password would have 6095689385410816 combinations!!

** Generally speaking, when trying to create a password, we are trying to create as entropic an outcome as possible as this will be the most computationally time consuming to break. The entropic value measured per key is calculated on the basis that each key press is independent and the entropy per key essentially increases with increased character range.

1 Source

2 Source

Due to the manner in which language is constructed, the occurrence of letter like vowels is dramatically increased leading to a much decreased entropy per key. This means, in order to create a reasonable secure 64bit key, you would need approximately 58 characters as opposed to only 10 if all characters are used.e

Vista SP1 - A reflection

It is interesting how perusing or glancing at the popular tech topic currently doing the rounds on wordpress can give an insight into the impact such announcements (or software/game/hardware/press releases etc) are having on the general public. Its all well and good reading about something (in this case Vista Service Pack 1) from recognised tech insiders such as Paul Thurrott, it is far more telling to read about the experiences everyone else is having.

Here are a selection of headlines from the last few days,

Vista wreaks havok on some PCs, users complain (anti Vista blog), My Nightmare trying to upgrade to SP1 (Insightful look into incompatible drivers), Vista SP1 update not showing up is for your own good (Reasons why SP1 may not be available yet for some people), SP1 Now available, Delayed, Delayed, Delayed, SP1 Day two (interesting positive feedback from a user), Hell has frozen over (overexcited user).

And guess what? Its not (entirely) the usual doom and gloom and has become almost ubiquitous when it comes to reports about Vista. Vista Service Pack 1 has come a long way since internal betas handed out to the Microsoft beta testers. These poor guys must have been feeling particularly abused this time round if the early write ups are anything to go by. With several restarts required to complete the process (and a few hours) these so-called tech elite reported back their thoughts on the process and as you can imagine, even the most staunchly pro-Microsoft of them has a few ‘choice’ comments to make.

But anyway, fast forward to now and you will see in your Windows Updates Vista Service Pack 1 waiting patiently for you to let it into your digital home. Software and hardware compatibility is good and the lengthy installation process has been slimmed down dramatically to a single reboot after completion. Inevitably it won’t go that way for everyone with some users reporting issues with certain drivers. To Microsoft’s credit, a fairly comprehensive list of drivers that have issues has been published and I have reproduced the list below.

Realtek AC’97

For x86-based computers: Alcxwdm.sys - version 6.0.1.6242 or earlier
For x64-based computers: Alcwdm64.sys - version 6.0.1.6242 or earlier
SigmaTel

For x86-based computers: Sthda.sys - version 5.10.5762.0 or earlier
For x64-based computers: Sthda64.sys - version 5.10.5762.0 or earlier
SigmaTel

For x86-based computers: Stwrt.sys - version 6.10.5511.0 or earlier
For x64-based computers: Stwrt64.sys - version 6.10.5511.0 or earlier
Creative Audigy

For x86-based and x64-based computers: Ctaud2k.sys - version 6.0.1.1242 or earlier
For x86-based computers: P17.sys – all versions (This was originally a Windows XP-based driver.)
Conexant HD Audio

For x86-based computers: Chdart.sys - version 4.32.0.0 or earlier
For x64-based computers: Chdart64.sys - version 4.32.0.0 or earlier
Display drivers

Intel Display

For x86-based computers: Igdkmd32.sys – versions between and including driver 7.14.10.1322 and 7.14.10.1403
For x64-based computers: Igdkmd64.sys – versions between and including driver 7.14.10.1322 and 7.14.10.1403

Unfortunately, I am the (not-so) proud owner of a Ac’97 soundcard in my primary laptop so it looks like I may have to fish around for drivers (AGAIN!!) to get my laptop to work properly with Redmond’s latest offering. The issues here are not Microsoft’s fault. Infact, as several tech insiders have noted, Microsoft was beating the drum about drivers to ODM/OEMs for months prior to Vista’s (and SP1’s) release but when the moment came to deliver, most manufacturers did not come to the party.

The reason is simply, it is not really cost effective. Take a computer you bought in the last few years (or Motherboard) and goto the manufacturer’s website and check the date of the ‘latest’ drivers (or BIOS.) Whilst these companies are fairly diligent during the product’s lifecycle, when they move onto something else, they stop putting out bugfixes or updates because it no longer makes commercial sense for them to pay their software engineers to do so. When Vista came out, many people had equipment (like me) that was designed for XP but could, with a fair amount of tweaking, run Vista very comfortably. The problem I (and many others) faced was a complete lack of native driver support for this hardware. I understand the problem, but I still think it is ridiculous. Microsoft did try to smooth this over by building in a compatibility layer into Vista to allow the loading of some XP drivers and while this helped a lot, there were performance penalties.

For now, I am not particularly fussed about SP1 so I will be sticking to vanilla Vista until either these driver issues are resolved (unlikely) or I get the time to find replacement drivers for my laptop.

MSN Spamharvesting - A Caution…

Let me set the scene, I was randomly on MSN when a friend of mine sent me a message saying “http://www.almanbirasi.info/list :)” On inspection, the site appears to offer a service to tell users who has blocked them on their contact list, a function not available in regular MSN.

I would highly advise anyone looking at any such ‘third party’ service to do a bit of background digging before giving ANY of their details over to a dubious third party. I was immediately suspicious and I will detail for you why.

1) The page looks well designed, but has no google page rank. This can mean one of many things, but usually it means a site or page is new. A service such as a MSN block checker is likely to spread virally (if it worked and was genuine) and this would likely lead to it being assigned a google page rank. An example where this SEO based tool can be very handy.

2) Whenever you are prompted for personal or account details, you should immediately stop for a sanity check - look to see who owns the site (can be done by a whois) and if your data will be secure. This takes on two forms, one - is the site reputable (can be checked a number of ways) and two - will your data be secure ? For the latter, considerations of encryption (is the page SSL secured) and data storage (does the site have a privacy policy etc)

3) The link ‘http://www.almanbirasi.info/list’ infact takes you to a page ‘http://www.msnliststatus.com/’ which is a bit odd.

4) The .info TLD - call me a domain-ist if you will, but I don’t trust this TLD.

5) The terms of service explicitly state that by using their service you are opting in to (a doubtlessly un-opt-out-able) advertising service.

As it turns out I was right, I sent a message back expressing my concerns to this person, assuming he had stumbled across it and sent me the link. Here is the conversation we had, edited to remove the name of the person.

Person A says:
http://www.almanbirasi.info/list
Konrad says:
what is this?
Konrad says:
haha no way!
Konrad says:
it has no page rank, i.e. its a great way to loose your login details to do something you can do within gaim / pidgin
Person A says:
huh?
Konrad says:
I would not use it
Person A says:
huh?
Konrad says:
looks like a big security hole
Person A says:
what?
Konrad says:
lol you tried it didnt you?
Person A says:
what?
Person A says:
did it just send a message for me?
Person A says:
fucking thing
Konrad says:
hahahaha!
Person A says:
sorry, just changing my password
Konrad says:
lol
Konrad says:
thats funny
Person A says:
did it send a message to you
Konrad says:
yes

So, as you can see, he used this utility and it logged into his account, sent me (and presumably others) a link back to its page without him either knowing or expressing his permission for this. It appears this has been going on for a few months with users also reporting having their accounts hijacked. Luckily there is a very simple way to determine whether you are on a person’s contact list or whether he/she has blocked you on a variety of accounts not just MSN like Googletalk, ICQ, AIM etc. It involves using a free open source program called Pidgin.

Install it, setup the account you want to use and login. Right click on the contact and select ‘Get Info’ this window should tell you everything you want to know about a certain contact as illustrated below.

My friend is very clued up about systems, IT and security but he fell for it, just be careful on the web.

Local Client Insecurity

Any tech-savvy user will know of a handful of security vulnerabilities relating to desktop computing, these can range from remote attacks (Man in the middle / Malware / DDoS / Brute Forcing / Port Scanning) to local exploits ( hardware & software keystroke logging / more Malware / dailers etc.)

In-fact, apart from the distant days of Windows 95 I cant recall a time when there were more things for security conscious users to be worried about. Back in the middle to late 90s, the internet was gradually becoming common place and within the reach of the layman. Unfortunately these users typically didn’t (or didn’t have a sufficiently fast connection - 4hrs for IE 4.0 update?) update software to patch security holes. The term script-kiddy was coined, referring to individuals who would use “off the shelf” exploit programs to wreak havoc. These easily found resources would be effective for months (if not years) due to the majority of users being completely clueless or disinterested in protecting their digital homes.

Fast forward to modern day, wireless hacking tools exploiting the poor design and implementation of WEP encryption have been commonplace for a number of years now. Wireless equipment manufacturers have taken on the role of securing their client’s networks by shipping routers with WEP (and more recently WPA) enabled by default which has helped secure many home networks from a variety of threats, from freeloading neighbours to network peeping toms. Security software companies have helped raise awareness while peddling their, often rather poor offerings to the unsuspecting public. (Norton anyone?)

By now everyone must know that running WEP on a Wifi connections is potentially extremely risky, those reading this who are still running an unencrypted Wireless Access point without some kind of secondary encryption system should stop what they are doing and read up on this.

It would appear that even wireless keyboards (using 27Mhz radio transmitters not Bluetooth) are vulnerable (although Bluetooth ones are also but via a different type of attack.) It turns out that security was probably very low down on the list of priorities during development of this common interface extension. The security system emplyed uses a single bit XOR encryption. The best explanation of how rubbish this is stems from TechFaq’s definition / explaination of XOR :

“XOR encryption is a trivially simple symmetric cipher which is used in many applications where security is not a defined requirement.”

The article concludes that there are only 256 possible keys that are set once a keyboard / receiver have been paired with no periodic shifting. It does make you wonder how easy it would be to build a portable device designed to record all 27Mhz data it can pull off the air for later analysis. Whilst the majority of the time it would capture useless keystrokes or harmless IM conversations, it could potentially capture bank details (although most banks now use secondary non input based authentication) or email / shopping account passwords. As if we didnt have enough to worry about with both software and hardware keyloggers already.

WD debuts DRM’d Harddrive (*hiss*)

DRM seems to be the popular industry buzz acronym at present regardless whether it inflame users or not. This recent trend of pandering to content providers over end users is both reprehensible and irresponsible from a corporate and moral prospective. A few weeks ago, Western Digital released on the unsuspecting world an external harddrive with buit in DRM.

I am a huge fan of Western Digital HDDs, and have bought 5 hard drives from them recently, however in such a competitive market, smart consumers have plenty of choice and will simply turn their noses up at a product that restricts their abilities. Wired posted that the SAMBA mode works properly, its only the client software that provides restrictions of Media files.

UPDATE: A whole community hell bent on replacing the official firmware with a F/OSS linux version has sprung up. Whilst I dislike products that force DRM on the consumer, I do love it when F/OSS projects are created on the back of them (e.g. iPOD linux)

Free (legal) Vista? .. how can I loose?

Microsoft have (quietly) started up a Feedback Program supported free version of Windows Vista Ultimate as well as other ‘popular’ software including Microsoft Office Ultimate 2007, Microsoft Money Premium, Microsoft Student with Encarta Premium 2008, or Microsoft Streets and Trips 2008. In order to qualify, you must agree to either one or both of the following programs:

“The survey feedback program. When you join the survey feedback program, you’ll be invited to take a survey on a regular schedule. If the survey arrives at a time where you are busy, you skip that one and take the next one instead. You will not receive more than a survey every two weeks.

The automated feedback program. When you choose to participate, most of the work is done behind the scenes, with no additional effort, time requirement, or inconvenience to you. Occasionally, we might send you an invitation to participate in a survey or another feedback program to get additional information about your use of computers but, in general, you only hear from us if we make significant changes to our data collection method or if we want to offer another feedback program to you that you might be interested in.”

Source

I have not had a chance to go through the program to see if there are any limitations other than periodic usage reports / surveys from Microsoft however if you are interested in upgrading to any other these pieces of software, it might be worth considering as a ‘demo.’ This idea is not a new one, versions of Microsoft Office have had voluntary opt-in usage statistics since Office 2000 as well as MSN Messenger. I don’t know about you, but the prospect of my operating system sending back ‘anonymous’ usage reports to a third party does not fill me with confidence.

MiniWin - A Windows Kernel in 4Mb

Whilst most of the details of the upcoming ‘Windows 7′ (or Blackcomb / Vienna depending on which codename MS are currently using) are still shrouded in Mystery. Microsoft recently revealed they intend to power Windows 7 with a brand new minimalistic core. At present Microsoft support a wide range of kernels across their products and it must be resource consuming maintaining all these codebases. After all, its not just desktop kernels but Mobile and PPC based systems as well:

‘Longhorn Based’ - Server 2008, Vista

‘Whistler Based’ - Server 2003, Windows XP, XP embedded, XP Media Center 2003/4/5, Tablet PC, XP Fundamentals, Windows Home Server

‘NT 5 / Asteroid Based’ - Windows 2000 Sever / Professional

‘Talisker Based’ - PPC OS, Smartphone OS

Magneto Based’ - Windows Mobile 5

‘Crossbow / Yamazaki Based’ - Windows Mobile 6

So the idea is to consolidate all supported architectures into one makes perfect sense in terms of longer term support. Of course, in order to be fit for purpose, the MiniKernel needs to be so flexible as to be used for embedded mobile applications as well as desktop AND enterprise grade server applications. The idea of a streamlined, optimised kernel is a welcome departing from Vista’s Bloat, I just hope Microsoft do not embed DRM into this mini kernel and bloat it unnecessarily.

The new kernel is purportedly to be 4Mb in size and uses only 100 files. Of course this precludes any GUI, however when run in a virtual machine with only 40Mb of RAM, the kernel (running a basic http server) used just 33 MB of the allocated RAM and booted in 20 seconds. This is a project fairly early in its life, and short-term will only make things worse (by adding yet another code-base) however it will be interesting to see what other optimisations can be made. Of course this kernel does not include all the parts needed for a fully fledged operations system like a graphical hardware layer and proper driver loading, however the implementation of a networking stack is promising. It shows what can be done when not implementing a broken IPv6 system that almost no-one presently uses..

It is one thing to show a prototype in an idealised virtual machine (ahem ReactOS :p) and quite another to provide an infrastructure that allows heterogeneous hardware support, however I am very interested in this new approach and like many others shall be watching this closely.

As a brief aside, I just want to say that my dig at ReactOS was precisely that, a playful dig. The project is one I have been following since 0.2.x and the work the guys are doing over there is extremely impressive, implementing an NT based F/OSS clone is alot of work (it too Microsoft ~15 years to get there after all!)