Wednesday, April 28, 2010

HellRTS,
The Not Quite In-The-Wild
Hacker Tool "Trojan"


"Take a deep breath and count to ten. One, two, three..."

This past couple weeks the incoherent, or should I say incompetent, nature of the anti-malware community has become evident yet-again through the discovery and discussion about a new variation of a hacker tool called 'OSX.HellRTS.D'. I am going to use that name because my best estimation is that it was first described by Intego, and that is their chosen name. It also follows the published malware naming protocol.

As per usual, other anti-malware companies could not bother to stick to the source name and have proliferated the usual WHATEVER of their own names, those being 'Hellraiser' and 'Pinhead'. Further confusion includes the addition of '.D' at the end of Intego's name. I have no idea why it is there. It indicates that this is supposed to be the FOURTH variant of this 'malware', and yet no one, including Intego of course, provides any reference to variants ".A", ".B" or ".C". I am going to toss out a wild guess that ".D" only means that this hacker tool has three previous versions known well only within the hacker community.

UPDATE: I have verified that 'Hellraiser' is the actual name of the source hacker tool of which HellRTS.D is the fourth variation.

Why I'm counting to ten: Because there are no signs of improvement in the chaotic nature of the anti-malware community. Anti-malware is supposed to be a 'professional' endeavor. The only reliably professional thing I have found so far in the community is that people make money in it. Otherwise, as a trained and experienced scientist, I find the community to be nothing more than 'A Pack Of Cards', as Lewis Carroll put it. That is to say it is a bunch of playing card characters disagreeing with one another over nonsense.


Which is to say quite bluntly:

The anti-malware community is not entirely scientific in nature.

(I am so itching to have someone disagree with my statement above. I dare you.)

Of course, enough of my own injection of subjective emotion into what should be an objective, scientific subject. Here's the lowdown on this new 'Trojan':

So far OSX.HellRTS is entirely ignorable. It is being distributed as a hacker tool out on the Internet, but has NOT been utilized as malware 'in-the-wild'. Instead it is being described as capable of being used as malware in-the-wild. When or if OSX.HellRTS becomes anything more than a hacker tool, I'll provide more detailed information.

In the meantime, here are some links for those who would like to dig around in the details:

David Harley has written a series of articles about HellRTS at his poorly named "Mac Virus" blog. David provides some very useful information through his professional work for the Mac community, which I very much appreciate. However, David also often makes his own contributions to the chaotic nature of the anti-malware community, fitting my fittingly harsh appraisal. Therefore, when you read his articles, "Take a deep breath and count to ten...."

Hellish Mac Malware

More on that hellish Mac malware...

OSX/HellRTS - more info

Here are Intego's source articles about OSX.HellRTS:

INTEGO SECURITY MEMO – April 16, 2010
HellRTS Backdoor Can Allow Malicious Remote Users to Control Macs


Intego Security Memo: HellRTS Backdoor Can Allow Malicious Remote Users to Control Macs

Now for the firing squad:

These are reports from other anti-malware companies that chose to use their own WHATEVER name for OSX.HellRTS. They should be lined up against a wall. As you click each of the links below, think to yourself:

"BOOM! HEADSHOT!"

Sophos: "OSX/Pinhead-B"

CA: "OSX/HellRTS"

iAntiVirus (PC Tools): "Backdoor.OSX.Hellraiser" <- Search on this page for 'Hellraiser' to read its description.

As usual I'm not going to bother with references to Symantec or MacScan articles. Why? Why bother.

If there are hackers who'd like to share the history of the Hellraiser hacking tool, please let us know via the comments! I'd be most interested.

Thursday, April 22, 2010

VLC Media Player Multiple Vulnerabilities
UPDATE!

UPDATE!

Hey kids. I found that in April some terrific folks on the Mac side of the VLC project have gotten things going again and have provided an update past VLC v1.0.5. You can download the lastest version of VLC at the source page for VLC media player Mac OS X Intel nightly builds. (Sorry PPC users, you are SOL).

Be sure to read the notes at the top of page very carefully! What you probably want is the latest version of the 1.0-branch-intel stable series. Ignore the gibberish numbers in the file names. When you see '107' in the name it does NOT mean 'version 1.0.7'. ATM the latest version is v1.0.6.
~~~~~~~~~~~~
Original Article:
~~~~~~~~~~~~


This is potentially a big problem:

VLC Media Player v1.0.5 (and earlier) has been found to have multiple vulnerabilities. And there is almost NO chance on a Mac OS X version 1.0.6 update at this time. Therefore, to be ultra-super-mondo-über-safe, you can't use VLC any more. And that's bad.

Here is Secunia's announcement.

You can subscribe to Secunia's Weekly Summary email newsletter HERE.

You can access the ongoing 'Vulnerability Report: Apple Macintosh OS X' HERE.

Now for the naughty bits:
Description
Some vulnerabilities have been reported in VLC Media Player, which can be exploited by malicious people to potentially compromise a user's system.

1) An error in the A/52 audio decoder can be exploited to cause a heap-based buffer overflow.

2) An error in the DTS audio decoder can be exploited to cause a heap-based buffer overflow.

3) An error in the MPEG audio decoder can be exploited to cause a heap-based buffer overflow.

4) An error in the AVI demuxer can be exploited to trigger an access to invalid memory.

5) An error in the ASF demuxer can be exploited to trigger an access to invalid memory.

6) An error in the Matroska demuxer can be exploited to trigger an access to invalid memory.

7) An error when processing XSPF playlists can be exploited to trigger an access to invalid memory.

8) An error in the ZIP implementation can be exploited to trigger an access to invalid memory.

9) An error in the RTMP implementation can be exploited to cause a heap-based buffer overflow.

Successful exploitation of the vulnerabilities may allow execution of arbitrary code, but requires that the user is tricked into opening a specially crafted file.

The vulnerabilities are reported in versions prior to 1.0.6.

Solution
Update to version 1.0.6.
Quick summary: Trojan horse files that are compatible with VLC can be used to access your computer's RAM and do bad things. Therefore, don't be a LUSER. Verify that any file you play in VLC is from a legitimate, reliable source.

For example, stuff from questionable BitTorrent sites does not qualify as safe. BitTorrent rips are a great way to get infected. That's where the one (and only) verified Mac BotNet came from: Three cracked Mac apps offered at BitTorrent sites turned out to be Trojan bots that zombied well over 10,000 Macs, employing them in a documented DDOS attack. The Trojans are formally known as Trojan.OSX.iServices.A-C. (You can read about this Trojan in previous posts).

So why no VLC v1.0.6 update for Mac?! Because there has been a long term derth of Mac OS X developers participating in the open source VLC project. You can read more about this dilemma HERE. You can sign up for the VLC RSS feed using this LINK. Of interest: There will be no v1.0.6 for Windows either, for different reasons.

Thankfully, there is an ongoing 'next generation' project of VLC for Mac called 'Lunettes' that you can follow HERE and HERE. This project is also asking for Mac developers. There is no public release version at this time.

What's so great about VLC? It allows Mac users to play a variety of media that Apple does not support via QuickTime or iTunes. Example: FLAC. This free, Open Source lossless compression audio format is ubiquitous on the Internet, and Apple provides no support, which is of course insane. There are many other examples of great functionalities provided by VLC. I consider VLC to be required software on my Macs. There are alternatives to VLC for some purposes, but no other application provides its feature set in one package. Let's hope the Lunettes project is a success.



For those concerned:

Please note that I post Mac-specific vulnerability information because just about no one else does. These posts help fend off the troll hate-abuse-lie rants and FUD mongering that claim that Mac users are computer security illiterate.

Wednesday, April 14, 2010

PDF Security Hole:
Hacking Into Copy & Print 'Locked' PDFs
-revised-

I was hoping I was wrong, but this is what I learned today:

Anyone can hack around a password required to copy from or print a PDF. Anyone.

Thankfully, the full locking of a PDF remains unhacked. The 'Open' password is still required.

A hacking tool that allows you to hack copy and print permissions is today's Mac Update Promo deal of the day. It is called PDFKey Pro. (48% off the regular price of $24.99). This program is a hacking tool that clearly points out a fundamental security hole in the PDF format. Therefore, I see no point in using PDF password for copy and print protection. It's worthless.

Please note that I am not knocking hacking tools. I am not knocking PDFKey Pro. The way it is being sold sounds entirely legitimate. The fact that copy and print PDF protections can be entirely defeated has nothing to do with the developer of this application. It has 100% to do with Adobe. Yeah, I know some people are tired of the onslaught of knocking against Adobe these days. Tough. This is a big fat and ugly nasty problem. Adobe are responsible.

In the past I've talked with both David Pogue and Adam Engst about selling electronic books. David Pogue and I talked about selling protected PDFs as one option. He decided at the time to try Adam Engst's method of simply trusting the customer. As an opponent of DRM (digital rights manglement), I agree with Adam. However, authors and publishers are entirely within their rights to prevent anyone from being able to copy from or print their documents and books.

Therefore, if you want to lock up the copy and print permissions of your docs, look elsewhere. There are plenty of great locking and encryption tools for Mac, but I'm not aware of anything that only prevents copying and printing.

I'd very much enjoy reading an analysis of how PDF protection is hacked. Something tells me it's already out there on the net for any hacker to read and use. What a shame.
--

Saturday, March 27, 2010

Tech Press Self-Immolation:
Blundered Pwn2Own Reporting

Tech press TechTardiness abounds. It is no surprise that certain dimwits blundered their reporting of the Pwn2Own contest at CanSecWest. My net compatriot Daniel Eran Dilger covered it laudably today in his article:

CanSecWest security competition falsely portrayed, again

Read and enjoy!

Thursday, March 25, 2010

64-bit 7ista Twice Hacked via both IE 8 and Firefox 3!
The End Is Nigh!


I should also mention that both Mac OS X 10.6 Snow Leopard and the iPhone got hacked via Safari. Just doing a little back-at-you priority swapping. These days it is a BIG DEAL when Mac OS X gets hacked because of its reputation as the safest GUI OS on the planet. Hacking Windows is ho hum because it happens every day.

Here are some links to somewhat detailed articles about the Day 1 results from the Pwn2Own contest at CanSecWest 2010 in Vancouver, Canada:

TippingPoint blog.
CNet.
MacWorld.

The contest still has two more days of hacking to go. But here is the current list of winners from Day 1:
PWNED! Vincenzo Iozzo and Ralf Philipp Weinmann - iPhone
PWNED! Charlie Miller - Safari [on Mac OS X 10.6]
Nils - Safari (Prize Claimed) [on Mac OS X 10.6]
PWNED! Peter Vreugdenhil - Internet Explorer 8 [on 7ista]
MemACCT - Internet Explorer 8 (Prize Claimed) [on 7ista]
Anonymous - Nokia
Anonymous - iPhone (Prize already won)
PWNED! Nils - Firefox [on 7ista]
Congratulations to all the hackers and thank you for making it clear that Internet surfing can be dangerous no matter the operating system or web browser. Details of each zero day hack are not published until they have been addressed by the companies or groups in charge of affected programs and operating systems. When the Mac OS X hacks have been published, I'll report them and provide links here.

I'll also post more from CanSecWest as it progresses. Dr. Charlie Miller will be presenting his 20 Mac OS X 10.6 Snow Leopard hacks.

The successful hacking of Windows 7ista is of particular interest because it involved bypassing the much lauded ASLR (Address Space Layout Randomization) and DEP (Data Execution Prevention) built into 7ista. So much for those security technologies!

In each hack the victim computers were directed to websites containing exploit code. I'm going to hazard a wild guess that the sites used code written at least in part in the catastrophic mess known as ECMAScript, aka JavaScript/JScript. Readers of this blog will already know my low opinion of this scripting language and my desire that it be banished from the Internet forever. Listeners to the SecurityNow Podcast know that Steve Gibson of Gibson Research Corporation (GRC) called out ECMAScript as dangerous years ago. He recommends surfing the net with scripting turned OFF in all web browsers by default, only turning it on at trusted websites.

Java exploits are also well known at this time, indicating the need to also turn off Java while surfing the net, except again at trusted websites. What a shame.

(Note that JavaScript and Java have nothing whatsoever to do with each other apart from a similar name caused by a marketing moron deal between Netscape and Sun Microsystems, both companies now defunct).

Monday, March 22, 2010

'Tis The Season For Pwn2Own!

--
FUD FUD FUD FUD FUD!
FUD FUD FUD FUD!













This is the time of year when, historically, anti-Apple security FUD is at its highest pitch. The great event begins March 24th. Our dubious hacking heroes Dr. Charlie Miller and Nils will be participating.
Pwn2Own 2010
BY AARON PORTNOY
MON 15 FEB 2010 16:41PM

The TippingPoint Zero Day Initiative (ZDI) is proud to announce that the annual Pwn2Own contest is back again this year at the CanSecWest security conference held in Vancouver, BC on March 24th 2010. As the contest name implies, if you successfully exploit a target you get to keep it along with a ZDI cash prize and related benefits. This is our 4th year running and to commemorate we have increased the total cash prize amount to $100,000 USD. If you're unfamiliar with the past history of this competition check out the archived 2008 and 2009 blog entries.
When the contest starts, you can follow the results at TippingPoint's blog HERE. The favorite to lose this year is Microsoft Internet Explorer, either or both versions 7 and 8. Here is the schedule posted by ZDNet:
Day 1:
Microsoft Internet Explorer 8 on Windows 7
Mozilla Firefox 3 on Windows 7
Google Chrome 4 on Windows 7
Apple Safari 4 on Mac OS X Snow Leopard

Day 2:
Microsoft Internet Explorer 7 on Windows Vista
Mozilla Firefox 3 on Windows Vista
Google Chrome 4 on Windows Vista
Apple Safari 4 on Mac OS X Snow Leopard

Day 3:
Microsoft Internet Explorer 7 on Windows XP
Mozilla Firefox 3 on Windows XP
Google Chrome 4 on Windows XP
Apple Safari 4 on Mac OS X Snow Leopard
ZDNet also reports that a number of mobile devices are part of a second set of hacking contests:
Apple iPhone 3GS
RIM Blackberry Bold 9700
A Nokia device running Symbian S60 (likely the E62)
A Motorola phone running Android (likely the Droid)
Apple, apparently in preparation for Pwn2Own, released Safari v4.0.5 on March 10, 2010. It patched 16 security vulnerabilities. You can read about it HERE and HERE. Six patches were specifically for the Windows version of Safari. The other ten patches affected both Mac and Windows versions of Safari. Nine of the patches were specifically for WebKit, which is an Open Source project used in a number of web browsers, including Safari, OmniWeb, Chrome, Shiira, Midori, S60, Android and the Palm Pre web browser. Four of the patches patched the ImageIO used in the version for Windows. Does this cover the gamut of security vulnerabilities in Safari? The hackers at Pwn2Own consistently have surprises up their sleeves.

You can read the details of this year's Pwn2Own contest HERE.

The general concept of the contest is to gather contestants and provide them with a hacking events schedule well ahead of time. The contestants typically come to the contest prepared with a specific hack or set of hacks they will use on the target computers via interaction with the accompanying web browser. This year the contest is somewhat different in that each successive day will include the hacking of older versions of Internet Explorer with older versions of Windows. But the general contest provides three days of hacking using three pairings of web browsers and operating systems. Day 1 does not allow any access to applications on the target computer. Day 2 allows what I call 'LUSER sabotage' access to the target computers via default installed applications for each operating system. Day 3 provides popular third party applications on each computer that can be used as part of 'LUSER sabotage' hacking.

In years past the FUD mongering contingent have danced around like village idiots pointing out how quickly Macs have been hacked on Day 2. In reality, the speed of any hack is nearly irrelevant. This is due to the weeks of preparation provided to all contestants, who presumably have already proven their zero day hacks before the contest has begun. What is relevant is the existence of the hack and how much 'LUSER sabotage' is required to apply it.

This year two senior contestants, Dr. Charlie Miller and Nils, will be using Safari v4.0.5 to hack into Mac OS X 10.6.2 Snow Leopard. Vincenzo Iozzo and Ralf Philipp Weinmann, as well as an 'anonymous' human, will be hacking into the iPhone.

One concern I have this year is that Safari is not being used to hack into any version of Windows. Instead only IE 7 & 8, FireFox 3 and Chrome 4 are being tested. Presumably the choices of Windows browsers were made according to market share as well as hacker interest. I'm also a bit annoyed that no Windows Mobile phones were included in the contests. Microsoft have announced the dumping of their current mobile OS for an entirely new mobile OS. But there is no reliable time line for this change, making the hackability of current Windows Mobile devices entirely relevant.

Hack and Enjoy!

Friday, March 19, 2010

Intego VirusBarrier Review Part II

--
My friend and former employer Michael Flaminio posted a very nice video review of Intego VirusBarrier version 10.6 over at Insanely-Great Mac. You can also access it at YouTube. I could not provide any improvement over Michael's review, so please give it a viewing! VirusBarrier is the only Mac OS X anti-malware program I can recommend for individual users. See my Part I review for further details and opinions regarding the program.

This past week someone told me that FUD mongers Symantec have finally gotten their act together, allowing their Norton Anti-Virus program to work properly without damaging your hard drive. Imagine that! I am seeking verification that this is indeed the case, if anyone would please let me know. Much obliged. If I get enough happy shiny smiley stories I may dare to perform some testing on the latest version myself.

Happy spring to the northern hemisphere! Happy fall to the southern hemisphere. (-_^)
--