07 July 2015

Data protection

Posts relating to the category tag "data protection" are listed below.

07 January 2015

Moonpig Website Vulnerability, Incident and Breaches

Personalised greetings card service Moonpig was all over the popular news yesterday.

Partial screen of the Moonpig customer support page that states 'A MESSAGE TO OUR CUSTOMERS: You may have seen reports this morning about our Apps and the security of customer details when shopping with Moonpig. We can assure our customers that all password and payment information is and has always been safe. The security of your shopping experience at Moonpig is extremely important to us and we are investigating the detail behind today's report as a priority. As a precaution, our Apps will be unavailable for a time whilst we conduct these investigations and we will work to resume a normal service as soon as possible. The desktop and mobile websites are unaffected.'

Paul Price found an exploitable weakness in Moonpig's public API and contacted them in August 2013, and again a year later. Eventually he gave up and published details on Monday.

Following much Twitter activity, yesterday Moonpig tweeted:

We are aware of claims re customer data and can confirm that all password and payment information is and has always been safe.

Interesting spin, since the vulnerability relates to other personal data — passwords or payment card holder data. Shortly afterwards, Moonpig tweeted:

As a precaution, our Apps will be unavailable for a time whilst we conduct these investigations: http://www.moonpig.com/uk/Information/Press/

Moonpig also added the following message to their customer support page:

A MESSAGE TO OUR CUSTOMERS: You may have seen reports this morning about our Apps and the security of customer details when shopping with Moonpig. We can assure our customers that all password and payment information is and has always been safe. The security of your shopping experience at Moonpig is extremely important to us and we are investigating the detail behind today's report as a priority. As a precaution, our Apps will be unavailable for a time whilst we conduct these investigations and we will work to resume a normal service as soon as possible. The desktop and mobile websites are unaffected.

Although Moonpig has not responded to the core issue (personal information), the published details appear to indicate:

  • Breach of principle 7 of the Data Protection Act
  • Breach of the Payment Card Industry Data Security Standard (PCI DSS)
  • A disregard for customers' data when the company has been aware of the problem for so long, and it continued to collect and process personal data through the period.

PCI DSS is only relevant here if the system components for api.moonpig.com are within the PCI environment. There is no need for a cardholder data breach for there to be a breach of compliance with PCI DSS. The main www.moonpig.com systems are definitely within scope since payment cardholder data is collected on forms generated by the website and the data is sent back to the same Moonpig website.

Nevertheless, by passing through the shopping basket and check out, other application security and privacy concerns are evident such as system information leakage, sending personal data over unencrypted channels, and third-party code on checkout pages.

The API issue and the other public issues on the web site do not seem to even meet the baseline security controls published for years by OWASP.

The help page about Payment and Personal Details Security states:

Security is an important priority for us and we are committed to protecting your privacy. We are registered as required under the Data Protection Act 1998 (Reg. Z4843659) and we use the most up-to-date technology available to protect your personal details. To avoid the risk of computer fraud, your credit card number is not stored in our system at any point in the payment process. Please see our privacy and security policy here.

That is clearly not true and might therefore be a breach of the Advertising Standards Authority Online Remit. The above also hints that somehow payment cardholder data is safe because it "is not stored". That's good, but it is not the same as saying it is not processed by Moonpig systems at all, which is likely to be misleading to some consumers. The terms and conditions say very little about protecting personal data - except from "in transit", and as we know that is not true for all parts of the web site that collect or display personal data.

If that is not enough for Moonpig, if the API vulnerability also affects United States customers, we will see the US Federal Trade Commission get involved. That body has been very strict in recent enforcement actions for online privacy failings. Affected US readers can submit complaints to the FTC online.

Posted on: 07 January 2015 at 12:55 hrs

Comments Comments (0) | Permalink | Send Send | Post to Twitter

07 December 2014

On Anonymity and Accountability

A post by information security practitioner Robert Hansen titled Anonymity or Accountability raised an interesting topic.

I wonder about the language here — "safety" and "freedom" are not opposites...

I think there is a terminology problem, and some misunderstandings about privacy from this security viewpoint. But a useful discussion to have. Read more on the WhiteHat Security blog.

Posted on: 07 December 2014 at 13:58 hrs

Comments Comments (0) | Permalink | Send Send | Post to Twitter

25 November 2014

Two ENISA Reports on Cryptography

At the end of last week, the European Union Agency for Network and Information Security (ENISA) published two reports on the use of cryptography.

One of the tables from ENISA's report 'Algorithms, Key Size and Parameters 2014'

Algorithms, Key Size and Parameters 2014 (PDF) provides guidance on appropriate cryptographic protective measures for the protection of personal data in online systems. The report defines primitives/schemes that can be considered for use today, as well as those for new/future systems. The document is intended for technical specialists designing and implementing cryptographic solutions..

The second report, Study on Cryptographic Protocols (PDF) extends the previous report to look at how the primitives/schemes are used in cryptographic protocols.

Both reports are free to access without registration.

Posted on: 25 November 2014 at 18:27 hrs

Comments Comments (0) | Permalink | Send Send | Post to Twitter

20 November 2014

TRUSTe Privacy Deception and Misrepresentation

US regulator Federal Trade Commission (FTC) has taken to task self-appointed privacy certifier TRUSTe (True Ultimate Standards Everywhere, Inc.) which labels itself as "powering trust and compliance".

Partial screen capture of a page from the TRUSTe web site showing some of its web privacy certification products

In a press release issued this week, the FTC states that TRUSTe has agreed to settle charges that it "deceived consumers about its recertification program for company's privacy practices, as well as perpetuated its misrepresentation as a non-profit entity".

Apart from a $200,000 fine, the proposed extensive settlement requires "TRUSTe will be prohibited from making misrepresentations about its certification process or timeline, as well as being barred from misrepresenting its corporate status or whether an entity participates in its program. In addition, TRUSTe must not provide other companies or entities the means to make misrepresentations about these facts, such as through incorrect or inaccurate model language.".

TRUSTe CEO Chris Babel's comment about the settlement can be found on the TRUSTe blog.

The United Kingdom TRUSTe web site is http://www.truste.co.uk/ which lists many UK clients. Sadly, a Google search indicates many of their clients haven't realised what's been going on and are still promoting the label.

Posted on: 20 November 2014 at 14:15 hrs

Comments Comments (0) | Permalink | Send Send | Post to Twitter

07 November 2014

75,000 GBP Fine For SQL Injection From ICO But With 90% Discount

Lancaster-based apartment booking company Worldview Limited has been fined under the Data Protection Act for allowing unauthorised access to customers' details. The company operates under two UK brands, Citybase Apartments and Central London Apartments.

Although customers' payment details had been encrypted, the means to decrypt the information - known as the decryption key - was stored with the data.

The Information Commissioner's Office (ICO) press release states that a SQL injection vulnerability that existed for 3 years was the root cause, so this might imply the the decryption key was either stored in the database or the database could be used to read the key from elsewhere, such as the file system. The information taken included 3,814 payment card details; this mentions that both primary account numbers (PANs) and three digit security codes were accessed, which is even more interesting. The terms and conditions (Citybase, Central London) state:

Your payment card details will be securely held for the purpose of processing the booking until the day of check in. On the day of check-in, the credit card details are removed from our systems.

That's the travel industry problem of stored card data.

Apparently the fine would have been £75,000 but this may have put the company out of business. However, I suspect the fact that Worldview Limited will also be paying forensic investigation charges, card re-issue fees, card monitoring fees and fines relating to their PCI DSS contractual obligations will also have been taken into account by the ICO. However, £7,500 is a lot less than Worldview should be spending to ensure their customer data is secure. The fine is reduced further to £6,000 if payment is made by 1st December 2014.

The monetary penalty notice is available on the ICO web site.

The two "site security" pages on both web sites (Citybase, Central London) put a lot of faith in the use of "industry standard Secure Socket Layer (SSL) encryption technology" only:

When you submit your card details the information is encrypted (scrambled) so that it can only be read by the secure server, making the transaction as secure as possible.

When Lush Cosmetics had an ecommerce incident in 2010-11 with a similar number of cards and other personal data compromised, there was no fine — just an undertaking (and of course the PCI DSS costs). I suspect this stronger response from the ICO reflects its view that SQL injection is a basic fault that is below any acceptable level of security.

Update 7th November 2014: Link to monetary penalty notice and details of early payment discount added.

Posted on: 07 November 2014 at 08:59 hrs

Comments Comments (0) | Permalink | Send Send | Post to Twitter

06 November 2014

OWASP Snakes and Ladders

In a month's time we will probably be in full office party season. I have been preparing something fun to share and use, that is an awareness document for application security risks and controls.

OWASP Snakes and Ladders Mobile Apps

Snakes and Ladders is a popular board game, with ancient provenance imported into Great Britain from Asia by the 19th century. The original game showed the effects of good and evil, or virtues and vices. In this OWASP version, the virtuous behaviours (ladders) are secure coding practices and the vices (snakes) are application security risks. I have created two versions so far:

I created the game to use as an ice-breaker in application security training, but it potentially has wider appeal simply as a promotional hand-out, and maybe also more usefully as learning materials for younger coders. To cover all of that, I use the phrase "OWASP Snakes and Ladders is meant to be used by software programmers, big and small".

OWASP Snakes and Ladders Web Applications

The game might be a useful transition from learning about the OWASP Top Ten Risks and before moving into the Top Ten Proactive Controls in a PCI DSS developer training session for example.

Snakes and Ladders Web Applications is available in German and Spanish, as well as in (British) English. Translations to Chinese, Dutch and Japanese are also in progress. The OWASP volunteers who are generously translating the text and performing proof reading are:

  • Manuel Lopez Arredondo
  • Tobias Gondrom
  • Martin Haslinger
  • Riotaro Okada
  • Ferdinand Vroom
  • Ivy Zhang

Print-ready PDFs have been published - these are poster sized A2 (international world-wide paper sizes). But the original files are Adobe Illustrator, so these are also available for anyone to use and improve upon. OWASP Snakes and Ladders is free to use. It is licensed under the Creative Commons Attribution-ShareAlike 3.0 license, so you can copy, distribute and transmit the work, and you can adapt it, and use it commercially, but all provided that you attribute the work and if you alter, transform, or build upon this work, you may distribute the resulting work only under the same or similar licence.

Just print out the sheet as large as you can make them. It is better to play using a real die and counters (markers), but you can cut out and make these from the paper sheet itself if you have scissor and glue skills.

You can also follow two mock games on Twitter which upload a position image every hour:

Please enjoy and share.

Further information, and all the PDFs and source files, are available on the Snakes and Ladders project website. Please keep in touch by joining the project mailing list.

Posted on: 06 November 2014 at 08:31 hrs

Comments Comments (0) | Permalink | Send Send | Post to Twitter

04 November 2014

Payment Checkout Flaws and Bugs

The announcement last week by researchers from Newcastle University about a problem with Visa's contactless cards reminded me to mention again commons issues with checkout and payment functions in web and mobile applications.

Photograph of customers in a household lighting stand during Clerkenwell Design Week 2014

The Visa fault relates to not enforcing the same limits on transactions when using foreign currencies. The paper is being presented this week at the 21st ACM Conference on Computer and Communications Security in Scottsdale, Arizona. While we hope we would not make similar mistakes ourselves, almost every web/mobile checkout/payment system I come across has some sort of problems.

I do not believe I have mentioned it previously, but if you are developing an online payment API, mobile or web payment application, you should read a paper from Microsoft Research issued in 2011. How to Shop for Free Online - Security Analysis of Cashier-as-a-Service Based Web Stores (presented at IEEE Symposium on Security & Privacy 2011 in Oakland, California) describes findings from research into the security of several web payment applications.

Many of these problems are data validation or authorisation issues, but can be labelled as "business logic flaws". My own checklist for reviewing payment application functionality is below:

  • Buy at arbitrary price
  • Buy at nil price
  • Buy without paying
  • Buy one at item at another item's price
  • Pay for one basket at another basket's price
  • Update the basket while paying for the original one
  • Voucher, gift card and discount enumeration or manipulation
  • Repeat order/payment
  • Missing "mandatory" steps
  • Refund after payment
  • Chargeback after payment
  • Pay customer instead of seller
  • Missing checks/enforcement of data validation/signing
  • Enumeration of accounts, customers, payment cards, baskets, orders, email addresses, phone numbers
  • Manipulation of out-of-band messages (e.g. emails, SMS, direct messaging)
  • Payment confirmation manipulation
  • Tax and currency conversion manipulation
  • Rate of use and floor limits
  • Staff/internal backdoors
  • Fraud opportunities
  • Test data/cards works/present
  • Third-party hosted content
  • Privacy contraventions
  • PCI DSS contraventions.

This does not describe every method, but I hope the list is of use to others anyway. Generic attacks (e.g. injection, path traversal, cross-site request forgery, man-in-the-middle, unpatched components) also crop up in ecommerce payment functions, like everywhere else.

Posted on: 04 November 2014 at 20:11 hrs

Comments Comments (0) | Permalink | Send Send | Post to Twitter

12 September 2014

ICO Seeks Feedback on Use of the Data Sharing Code of Practice

Further to my post on Monday about the new privacy seals consultation, the ICO has requested feedback on the use of one of its major guidance documents.

Photograph of the transparent cased Sinclair ZX-80 computer exhibit at the recent Barbican 'Digital Revolution' exhibition in London

The Data Sharing Code of Practice was launched in May 2011 and provides statutory guidance on all internal and external sharing of personal data. The ICO has requested feedback on how the code is being used by organisations, in the form of an online survey

The survey runs until 5th October 2014.

Posted on: 12 September 2014 at 08:11 hrs

Comments Comments (0) | Permalink | Send Send | Post to Twitter

08 September 2014

ICO Consultation on Privacy Seals Endorsement

The UK's data protection agency Information Commissioner's Office (ICO) has published a consultation on privacy seals.

The ICO will endorse at least one scheme for a minimum of three years and will review all endorsed schemes in the final year.

The privacy seals consultation comes closely on the heals of BSI's new application security kitemark, but is requesting feedback on the concept of endorsing privacy seals schemes.

The Framework criteria for an ICO-endorsed privacy seal scheme describes that the ICO's endorsement is conditional on the privacy seal scheme's operator achieving official accreditation by the UK Accreditation Service (UKAS). The ICO will invite proposals for a privacy seal scheme in autumn, with a view to selecting a provider early next year, and . launching the first round of endorsed schemes in 2016.

A pre-formatted consultation response document has been provided that can be returned to the ICO by email or post.

The consultation closes on 3rd October 2014.

Posted on: 08 September 2014 at 07:19 hrs

Comments Comments (0) | Permalink | Send Send | Post to Twitter

19 August 2014

SSL-Only and This & That

It should come as no surprise that Google has indicated it has begun noting the use of HTTPS (HTTP over SSL) on some web sites to contribute to ranking in its search results.

we'd like to encourage all website owners to switch from HTTP to HTTPS

The writing is on the wall for almost any site that is not SSL-only. I have worked with some organisations that have shifted to SSL-only to simplify session management protection and to provide some identity assurance. The often-mentioned processing overhead has never been an issue for customers or server-side. There is plenty of discussion about Google's announcement (such as here, here and here).

A few tips from the trenches if you are about to renew or buy a new SSL certificate:

  • Use 2048-bit key certificates
  • Ensure your certificate provider signs it using SHA-2 (not SHA-1 which will be deprecated at the end of 2016); even if you ask for this it may not be what you receive — check it (but note pre Microsoft Windows XP SP3, any outdated web browsers do not support SHA-2 signed certificates)
  • If you need multiple domains within the same certificate, but not a wildcard certificate, be very careful about certificate product selection because, while any number of domains might be supported, this is often used to differentiate products and thus cost
  • If anyone mentions elliptic curve cryptography (ECC) digital certificates or hybrid ECC, probably ignore them for now (unless you are a bank)
  • Make sure you set the HTTP Strict Transport Security header and never include any non-SSL content in pages (often a problem if third-party hosted content is included).

If you want to delve deep into the protocol or need more implementation guidance, I can highly recommend Ivan Ristic's newly completed book Bulletproof SSL and TLS. It is written in a quite engaging style and is very readable in a technical sort of way (the plot is a bit weak though). And, if your website is publicly accessible, his server testing tool at SSL Labs is probably the most comprehensive and easy to use method for verifying the configuration. If you want a local tool, look at Achim Hoffmann's O-Saft.

My most relevant previous related posts on SSL are:

Oh, and beware public shaming where SSL really ought to be in use already. Some regulators and organisations you may have contractual obligations to might comment too.

Update 23rd August 2014: See the post about HTTP Strict Transport Security (HSTS).

Update 7th September 2014: See also Google's announcement about support for SHA-1 SSL certificates in Chrome.

Posted on: 19 August 2014 at 19:16 hrs

Comments Comments (0) | Permalink | Send Send | Post to Twitter

More Entries

Data protection : Application Security and Privacy
ISO/IEC 18004:2006 QR code for https://clerkendweller.uk

Requested by on Friday, 27 November 2015 at 08:08 hrs (London date/time)

Please read our terms of use and obtain professional advice before undertaking any actions based on the opinions, suggestions and generic guidance presented here. Your organisation's situation will be unique and all practices and controls need to be assessed with consideration of your own business context.

Terms of use https://www.clerkendweller.uk/page/terms
Privacy statement https://www.clerkendweller.uk/page/privacy
© 2008-2015 clerkendweller.uk