Posts tagged as:

General

Network Security Podcast, Episode 272 v2

by netsecpodcast@mckeay.net (Martin McKeay) on April 24, 2012

in SBN

As a follow up to last week’s episode, Martin was joined last week by Josh Corman to talk to Wade Baker about the 2012 Verizon Data Breach Investigation Report.  Wade talks to us about how the information for the report was gathered, some of the strengths and weaknesses of the analysis and finally how the amazing puzzle that is the front cover was concieved.  The episode is a little longer than normal, but worth the time.

When this podcast was first release, it was mistakenly seen by iTunes as the PDF of the DBIR as being the podcast.  Subsequent attempts to upload were similarly misidentified.  Here’s hoping that a remix of the podcast will be significantly different enough that it doesn’t try keying on the DBIR again

Network Security Podcast, Episode 272v2

[Slashdot] [Digg] [Reddit] [del.icio.us] [Facebook] [Technorati] [Google] [StumbleUpon]

What do printed QR codes and NFC (Near Field Communication) chips have in common, besides storing instructions that computers can read? They are both hackable and their ability to store and communicate computer instructions is bound to be abused, if not already, then sometime soon. This happens to every new means of communication; QR and NFC are no exception. Call it "Cobb's first law of communications abuse" or just a statement of the obvious: Every new means of communication will be abused. Of course, the second law states that the abuse will include, if at all possible, the spreading of malicious software.

An example of a QR codeThis blog post is not about how to abuse QR codes or NFC chips–sooner or later people are going to figure that out for themselves–we just want to take a moment to urge companies and coders working with these technologies to implement them as sensibly and securely as possible. Failure to heed this warning could mean cool technology being crippled by clumsy protection schemes bolted on to fix security issues created by naive implementations. 

Consider the problem illustrated in the following video called "Card Tricks". This shows some rather worrying behavior by a pair of fairly common smartphones–an iPhone and a Samsung Nexus phone–being used in default configuration, no hacking required. It should be clear that a malicious party can go several steps further with this technology to create a means of infecting mobile devices, stealing data, and accruing ill-gotten gains from bogus phone charges and other scams.

The problem we see in both of the examples–the QR code scanning by the iPhone and the NFC tag reading by the Samsung smartphone–is that the software which interacts with the code/tag proceeds to act on the data in the code/tag without asking permission. While the desire for immediate gratification is understandable, and the ability to deliver it with technology is very cool, I am hoping the video makes it clear why this is not always a good thing.

QR and NFC software should not act upon the instructions in a code or chip without providing the user with an informed choice. A preview-and-authorize process should be the default mode of operation. Otherwise the technology becomes a high tech version of an old party trick: A person wearing a blindfold is told to put their hand into a jar to grab a prize, but what they grab might not be a prize at all.

AT&T QR code scanner/readerRight now there don't seem to be many people exploiting the "instant gratification" mode of QR and NFC scanners for evil purposes (although some cases have been reported). So you might ask: Why talk about the problem now?

My hope is that we can get ahead of the curve in terms of user education by convincing people to exercise caution when they use this technology. For example, if you use an iPhone you know it doesn't come with a QR code reader installed. Choose one that has a preview-and-authorize mode and set that as the default.

The QR app that I installed for the video demonstration–called QR Reader for iPhone–was picked at random from the app store. It is a free download, made by TapMedia and has an almost 5 star rating, so it is reasonable to assume that a lot of people are using it. This app does have a preview and authorize mode but that is not the default setting.

The QR scanner that AT&T offers, also free and a solid 4 star rating, has preview-and-authorize set as the default mode of operation which earns AT&T bonus points in my opinion. As NFC chip readers become more widely deployed in smartphones I'm hoping all of the associated software will have preview-and-authorize as the default setting. (This does not appear to be the case on the Samsung Nexus that I tested.)

Another reason to talk about this problem now, rather than later, is that a lot of of QR codes are showing up in stores and a ton of new NFC-enabled smartphones will be hitting the streets this year, opening up the possibilities for those who would seek to abuse them. On the Android platform I have already noticed "Control Near field Communication: Full internet Access" as one of the permissions sought by mainstream apps like Google Earth and YouTube.

A further aspect of the emerging NFC story will be phones like the Sony Xperia S. Announced with much fanfare a few months ago, this product features matching NFC tags that allow you to alter the settings on the phone, as illustrated in this frankly scary Sony video. Even if you give Sony the benefit of the doubt and assume that the Xperia tags are somehow bound to each phone by unique identifiers, it is a fair bet that before long someone will announce that they can craft which are able to change the settings on other people's phones.

Another fair bet is that one or more of the following will happen before the end of the year:

  • Pirate NFC chips: Will be deployed to divert traffic or execute other unexpected instructions in order to compromise smartphones.
  • Pirate QR code stickers: As foreseen last year by my colleague Cameron Camp, people will try sticking their own QR codes over existing codes to divert traffic or execute other unexpected instructions.
  • QR code seeding: People will include malicious QR codes in cheap flyers and distribute them to find and target victims.
  • QR Sharpie hacking: People will figure out how to alter public QR codes with a Sharpie to break the code or divert traffic.

Although QR codes and NFC chips are technically quite different they suffer the same potential weakness: Trusting that the person who programmed them has good intentions. I see them as a new attack vector for mobile devices and another reason to protect those devices with appropriate malicious code, app, and URL detection.

Finally, to the last piece of the card trick: How do people make QR codes and encode NFC tags. The software to create QR codes is readily available on the web and there are a lot of websites that will generate the code for you (if you need QR codes for a legitimate project I strongly suggest you test the code before deploying). As for NFC tags, I will go into this in a separate blog post about Near Field Communications that will appear in the near future, but the short version is that acquiring and encoding NFC chips is cheap and easy. Stay tuned.

AVAST gets Advanced Plus rating in AV-Comparatives’ Test

by Deborah Salmi on April 23, 2012

in SBN

Avast! Free Antivirus 7 has the distinction of being the only free antivirus to receive the Advanced Plus certification rating from the annual “On-Demand Detection of Malicious Software” test from Anti-Virus Comparatives.

Approximately 300,000 pieces of malware were used in the testing, and avast! Free Antivirus 7 detected 98% of them; the highest detection rate of all tested free solutions which outperformed a number of paid-for products from other AV vendors. Complementing the high malware detection rate, avast! was also recognized for detecting few false positives during the test. The number of avast! false alarms was 14. The average was 48 false positives. Avast! Free Antivirus 7 is the only free antivirus to receive the Advanced Plus certification rating.

AV-Comparatives chooses which antivirus products are to be tested from a field of internationally well-known, up-to-date antivirus products. In order to ensure that test results give a complete and accurate picture of a product’s capabilities, AV-Comparatives has strict rules about which tests every product must take part in, and which tests are optional. A dynamic “real world” protection test is conducted which measures file-detection rates, the number of false positive alerts raised, as well as other tests that cover different features of the products.

TEOTWAWKI … Take 2

by C. Warren Axelrod on April 23, 2012

in SBN

Do you remember all the Doomsday folks setting up shelters and supplies to last through the Y2K meltdown … the one that didn’t happen? Well, now we have another group, this time consisting of four individuals, each willing to spend $7 million in preparation for the EMP (electromagnetic pulse) catastrophe that they expect from sunspot eruptions … see “Doomsday set prepares for end in luxurious missile silo” at http://now.msn.com/now/0409-luxury-condos-in-abandoned-missile-silo.aspx They must have read William Forstchen’s book, One Second After, with its foreword by Newt Gingrich, which I discussed in a couple of earlier BlogInfoSec columns. The book relates the total disaster resulting from a purposeful EMP attack versus one emanating from natural causes, although the outcomes could be comparable.

In any event, I was reminded that I had worked on a column on the topic a few months back, but hadn’t published it. The title of the column was “EMP Now a Political Issue” and it went like this:

I was somewhat surprised to see in a front-page article, “Among Gingrich’s Passions, A Doomsday Vision,” in the  December 12, 2011 New York Times  by William J. Broad, that Republican presidential candidate Newt Gingrich is raising the issue of the potential impact of an electromagnetic pulse (EMP) attack to the top of his agenda … see http://www.nytimes.com/2011/12/12/us/politics/gingrichs-electromagnetic-pulse-warning-has-skeptics.html?_r=1&pagewanted=all  I wouldn’t have been so surprised had I remembered that Gingrich wrote the foreword to the William R. Forstchen book, One Second After, which I discussed two years ago in my December 28, 2009 column, “EMP (Electromagnetic Pulse) – Yet Another Critical Infrastructure Concern.”

(...)
Read the rest of TEOTWAWKI … Take 2 (419 words)

© BlogInfoSec.com, 2012. | Permalink | No comment | Add to del.icio.us
Post tags: , , , , ,

DNS Changer (re)lived, new deadline: 9 July 2012!

by Righard Zwienenberg on April 21, 2012

in SBN

As written in our “Password management for non-obvious accounts” blog post on 22 February 2012, the FBI confiscated the DNS Servers used by the DNS Changer malware and replaced them with different servers so that infected users would not be left without internet right away. Initially these replacement DNS Servers were to be taken offline on 7 March 2012. That meant any users who had not yet cleaned their systems and restored their DNS configuration would find themselves unable to use the internet.

All internet users were urged to check and clean their systems before 7 March. However, due to the large number of affected systems and the unprecendent amount of effort involved, a federal judge in New York has ordered that the replacement DNS Servers not be taken offline before 9 July 2012. This may seem like good news but in reality it means too many systems are still affected and dismantling the replacement DNS Severs would cause havoc.

One way to check if your system is affected by this DNS Changer malware is to use a free DNS check that several websites offer. If you rely (or have relied) on these websites, please be aware that some websites may be tampered with or malicious and give you the wrong advice. Just to be sure you use correct websites that offer this feature, ESET has verified that this the US-website and this European-website have the proper checks and give sound advice.

Another good way to check if your system if affected by any variant of the DNS Changer malware or any type of other malware is to use the free ESET Online Scanner. ESET’s Online Scanner employs ThreatSense scanning technology and is updated several times a day with detection and automatic remediation of newly discovered threats.

Please be aware that if your system is still affected by DNS Changer come 9 July you may not be able to use the internet any longer. When that happens, especially for less technical people, it may be problematic to clean your system manually. A quick online scan just takes a short time and can prevent you and your system suffering further complications that could be awkward to resolve.

Something to think on from Source Boston

by netsecpodcast@mckeay.net (Martin McKeay) on April 18, 2012

in SBN

“The Internet will never again be as free as it is this morning” – Dan Geer at SOURCE Boston

Think on that for a while.  If it doesn’t scare you, it should.

[Slashdot] [Digg] [Reddit] [del.icio.us] [Facebook] [Technorati] [Google] [StumbleUpon]

Phishing Using HTML and Intranet Security Settings

by Righard Zwienenberg on April 16, 2012

in SBN

Phishers always try to find new ways to bypass security features and trick ‘educated’ users. Over the years we have seen simplistic phishing attempts where the required information had to be typed into the e-mail body. This worked at that time because phishing was new and hardly anyone had a notion of the implications. Later, when spam filters became aware of these kinds of mails, we saw the evolution to direct links in e-mail, then to obfuscated links in e-mail where the e-mails looked professional and had the appearance of official messages from the organization the phishers desire your information from.

One thing stayed the same and that is the language used in the phishing mails. Most often they are not correct in either a contextual way or grammar-wise. Regardless, all these attempts sooner or later will be blocked by spam filters or by the anti-malware products, or by URL reputation schemes such as Google’s Safe Browsing or Microsoft’s Smart Screen. But, too often, some people still fall victim for these phishers. One observation as to why this happens will be described later in this blog.

In the last few weeks, a new approach can be added to the portfolio of phishing attempts. The e-mail is accompanied by an attachment with the extension “htm” or “html”.

Now why would the phishers use an attachment with html-code rather than a link? There are several reasons for this. Amongst others:

  1. Executing the html-code locally will not have your browser go to a website and thus the URL reputation filter will not be applied.
  2. Executing the html-code locally for the browser means it is loaded from an intranet rather than from the Internet. Settings are usually less strict for files started/loaded from an intranet.

Some applications have a protection mechanism in place when specific files are executed or loaded directly as an attachment as, for example, Microsoft Office 2010, which will only load the file in a Protected View where active content is disabled

If we look at this specific phishing attempt, besides the fact that the window is not really a properly delineated window with a border, the grammar is far from perfect.  Also ‘PayPal’ is addressing the recipient as “Dear Paypal Member” where the real PayPal (and other legitimate financial organizations – at least, those with a clue) will always address their customers by name and/or some other unique identifier known to both the provider and the customer. Also interesting is that this e-mail supposedly is coming from PayPal France, but the message is in English.

It implies that they already have information and they want to confirm it. Just double-clicking on the attachment will result in a warning.

We have already started to see e-mails where the reader is instructed to save the e-mail to the local system first, often to the Desktop for convenience, and then to execute or load them from there. But when the files are executed or loaded from the local system, the last safety-net provided by the mail program is bypassed.

With this phishing attempt we will see the page, presumably with the information they claim they already have and want to verify.

Interestingly they seem not to have any information from their customer yet. wink

Also noteworthy is the button to submit the information. Where the e-mail supposedly came from PayPal France, and the text and information is all in English, the submit-button has the text in German.

When the information is submitted, the form will contact a website where the information is retrieved using a php-script. When that is finished, the user will be forwarded to PayPal’s American homepage , resulting in a rather interesting collection of different languages in this phishing attempt.

So why do people still fall for phishing attempts? Education is the answer! Most people are completely new to the topic. If we limit ourselves to phishing for bank information, we know that more and more people move towards banking-online. Either for convenience, or driven by the fees the banks charge to bank in the old way, or because their bank has stopped allowing traditional banking methods. Lots of older people suddenly need to go online to do their banking and are unaware of these phishing attempts, and so fall for them.

It would be a good thing for banks to generate solid documentation on what a bank will NEVER ask for by e-mail, on their website, by telephone or any other means. It is in the banks’ own interest to educate their new online customers, as we expect that in 2012 we will see more and more of these phishing attempts.

ESET researchers have generated quite a few documents on phishing, and while some of them have been around for a while, the basic principles and advice on how to spot phishing emails still hold.

A Pretty Kettle of Phish by David Harley and Andrew Lee
Understand and avoid the attentions of phishers and other Internet scammers.

The Spam-ish Inquisition by David Harley and Andrew Lee
A detailed overview of spam, scams and related nuisances, and some of the ways of dealing with them.

Phish Phodder: Is User Education Helping or Hindering by David Harley and Andrew Lee
Evaluates research on susceptibility to phishing attacks, and looks at web-based educational resources such as phishing quizzes. Do phished institutions and security vendors promote a culture of dependence that discourages computer users from helping themselves?

We’ve also written many blog articles on the topic.

 

The Case Against Unindicted IT Co-Conspirators

by Michael Vizard on April 16, 2012

in SBN

One of the things that is generally assumed about IT security is that the bad guys are making use of some massive amount of IT infrastructure they invested in to send malware. In reality, a report from Agari, a provider of an anti-phishing cloud security service, suggests that real primary source of all that malware out there is systems that have been compromised.

Fighting the OSX/Flashback Hydra

by Pierre-Marc Bureau on April 13, 2012

in SBN

The biggest Mac botnet ever encountered, the OSX/Flashback botnet, is being hit hard. On April 12th, Apple released a third Java update (http://support.apple.com/kb/HT5247) since the Flashback outbreak. This update includes a new tool called MRT (Malware Removal Tool) which allows Apple to quickly push malware removal code to their user base. The first mission of MRT: remove Flashback.

A lot of researchers and security companies have been interested in OSX/Flashback. Many have published observations and partial results, generating a lot of buzz. ESET has been actively investigating the OSX/Flashback botnet. ESET was one of the first companies to implement a sinkhole to monitor the botnet. We can confirm the magnitude of the infection spread reported by other companies: we have seen more than 491,793 unique IDs coming from over 749,113 unique IP addresses connecting to our sinkhole. We are actively collaborating with the security community, sharing the results of our reverse engineering efforts and sinkhole data.

The OSX/Flashback malware can infect computers by multiple means. In the last couple of months, we have seen it spread as a fake Adobe Flash player (hence its name) and through exploits. The bulk of the infections happened recently when a group of websites started distributing the malware through drive-by download, exploiting the CVE-2012-0507 vulnerability in Java.

The first stage component of OSX/Flashback is a dropper, its only functionality is to contact a command and control server, download additional components and run them. Some of the variants of the dropper we have seen would also load a library. When installed, the library will load with any application on the system. It hooks the system functions responsible for communication and is in a position to alter web pages and spy on users’ internet activity and behaviour. It is still unclear to us if this spying is used to display unsolicited advertisements in the browser of infected computers or to steal information.

When it comes to disclosing a realistic number of unique infected hosts, we strive to be as accurate and objective as possible. Defining a unique host is not trivial, even if OSX/Flashback uses hardware UUIDs. Our data indicates many UUIDs that connected to our sinkhole (a server we set up to capture incoming traffic from bot-infected machines trying to communicate with their command-and-control servers), came from a big range of IP addresses, indicating that there may be UUID duplicates. Virtual Machines or so-called Hack-intosh installations may explain this.

When browsing Hack-intosh forums, we found out that everyone who is using the fourth release candidate of a special distribution has the same hardware UUID (XXXXXXXX-C304-556B-A442-960AB835CB5D) and even discuss ways to arbitrarily modify it.

Funny enough, we found this UUID connected to our sinkhole from 20 different IP addresses. This indicates that those who considered UUID to count the number of distinct infected hosts probably have underestimated the botnet size.

Flashback evolved a lot in the last few months. The authors moved fast and added obfuscation and fallback methods in case the main C&C server is taken down. The dropper now generates 5 domain names per day and tries to get an executable file from those websites. The latest variants of the dropper and the library encrypt its important strings with the Mac hardware UUID. This makes it difficult for researchers to analyze a variant reported by a customer if they don’t also have access to the UUID.

The fallback mechanism that Flashback uses when it is unable to contact its C&C servers is quite interesting. Each day, it will generate a new Twitter hashtag and search for any tweet containing that hashtag. A new C&C address can be provided to an infected system this way. Intego reported this last month, but the latest version uses new strings. Twitter has been notified of the new hashtags and are working on remediations to make sure the operator of the botnet cannot take back control of his botnet through Twitter.

To protect you Mac OS X computers we highly recommend applying the latest update from Apple. In addition, users can also download a (free) trial version of ESET Antirivus to scan their computer for infection and clean any threat that might be found on the system.

Thanks to Marc-Etienne Léveillé and Alexis Dorais-Joncas for their contribution to this research.

Pierre-Marc Bureau
Security Intelligence Program Manager

Pinterest security update

by Cameron Camp on April 12, 2012

in SBN

We recently highlighted a security walkthrough on Pinterest.com, the pinboard style sharing website that’s taking the social media by storm. Since then, they’ve continued to grow, and continued to have accompanying growing pains common in organizations with rapid growth. Here we highlight ways they are adapting, changes they are making, and what it means to you.

First, we note that Pinterest, by one account, drives more referral traffic than Twitter, no small feat. We also read that traffic spiked 52 percent between January and February, from 11.7 million unique visitors to 17.8 million, according to a comScore report. On its meteoric rise, it has faced issues ranging from copyright problems to fake gift card scams, and now we are seeing cybercrooks focus squarely on the platform as a delivery method for their scams to potential new/unfamiliar audiences.

The gift card scams start by purporting to offer free goods or services, ranging from coffee gift cards to free iPads. We’ve seen this before with more traditional web-based scams, but here the scam is tailored to Pinterest, coaxing the user to click on the pinned entry and visit endless survey websites before getting the alleged gift card. The twist is that scammers add a step required to “get your free gift card” that includes you re-pinning the original scam, thereby spreading it in your name, seeming to be coming from you instead of the original scammer. From there, some users are encouraged as a final step before getting the gift card, to install software, which would guarantee a steady supply of pop-up ads and other potentially unwanted applications, or worse. While Pinterest has attempted to crack down on these scams, and users become familiar to them and get wise, still the scams are propagating.

Then, there is the issue of copyright. While not strictly a security issue, still users could become exposed to potential violation of copyright of a given work, to the chagrin of more than a few users. It seems that a user is expected to comply with the copyright of a photo they post, for example. But what happens when that same image gets re-pinned, possibly extending its exposure far beyond the scope of the original copyright, a burden which the old terms of service attempted to place on the original poster? That (and other related) policy has been updated with the recently release updated Terms of Service, which you can read here.

Now we see Pinterest has produced an API interface for other apps to interact with the service, so we’ll wait and see if this exposes new security risks or exploits. To address this, other services have enlisted a paid bounty program to reward researchers for finding and reporting issues rather than exploit them, which seems to be effective at Facebook and Google for some time now. Hopefully Pinterest will consider some such program, or crowd-sourced variations, which will beef up the number of security specialists watching for problems – hopefully before they happen.

In the meantime, many users have been caught off guard by the amount of their Facebook information (since you are required to use either Facebook or Twitter account to sign up for Pinterest) which seems to “magically” appear on Pinterest, when they login to the site, especially pins from users whose names are familiar – from the Facebook friend list. One way to ensure that a minimum of information is cross-shared (if you are predisposed to restrict it for security reasons, to protect data sprawl, or otherwise) is to restrict your sharing settings in you Pinterest settings page. By ratcheting these down, you can exercise more control over what portion of your friends’ information that may ooze over to Pinterest, for uses they see fit.

We’ll continue to keep an eye on the security stance of the service as it continues to expand. But the usual advice applies: watch for offers that look “too good to be true”, and use a more minimalist approach to sharing and cross-sharing across your friends/contacts from various social media. You’ll be glad you did, and so will your friends, whose information may be more well-protected against data sprawl, and its accompanying problems.