There’s a rumor going around and a lot of unconfirmed information that a facebook application known as fan check is actually a virus. I’m seeing several claims that if someone becomes a fan of this facebook app (which the page claims is undergoing changes and currently unavailable)… well if someone becomes a fan in the meantime their friends will all be notified of the app, password on the account changed, etc. I’m a bit skeptical at this point because EVERYTHING I see seems 3rd party and I don’t see anyone claiming that THEIR account was compromised. What I DO know is that some of the top search results are known malware attack sites. For instance the second result (which was the top result a few hours ago….

Is eaves.ca/cgi/facebook-fan-check-virus.htm and the page it redirects to pushes the Personal Antivirus download. I don’t know if you’re familiar with Personal Antivirus but it’s one of those rogue antivirus applications that tries to push you into downloading and then buying with all sorts of trumped up virus and problem claims.

So, I don’t know if the fan check virus is legit or not, what I do know is that you need to be cautious with the search results. If you get something pop up offering to scan your pc for viruses close the browser window. Don’t download the scareware that they’re pushing and for heavens sake don’t pay a cent to the companies that push these scare and other slimy tactics.

I’ve seen a couple of these emails today and wanted to give a post just to warn people that these are bogus and you should NOT follow the link suggested in the email. I HOPE no one reading this falls for it, but the “tax software update” that they are pushing is a virus. (SHOCK!) Only a little over half the antivirus vendors currently detect it.

Read on for details on the message body…

I ran it through virustotal and it’s a variant of mytob according to some antivirus vendors.

Here’s the body:

Dear Tax Payer,

As part of new requirements from the IRS, all U.S. Citizens are required by law to update their computers with new tax software.

To begin the update, please visit http://65.15951047 and click “Open” when asked how to begin the download.

After doing so, no further action is required on your part.

Thank you for your cooperation,
IRS.GOV Agent #7[3

=======================

The only variation in the text between messages seems to be the last line...

IRS.GOV Agent #0[3

is what I saw in another message.

Both messages seem to be from the same machine... here's the initial received header.

Received: from Exploit ([92.48.88.145]) by domainremoved (8.13.1/8.13.1) with SMTP id m24LIbv9002684 for
; Tue, 4 Mar 2008 14:18:39 -0700

Gee, looks like a cool uberhacker calling their machine “Exploit” —better look out for them….

Sender addresses seem to be quasi-random… name+2-3numbers@irs.org (I wonder why they didn’t just try to spoof irs.gov?)

The address should not be visited obviously without the biohazard suit…, it contains a file program.exe served up in an frameset which means that on visiting the page there is a file popup to download/run.

The http address resolves to a machine at ip address 65.243.100.199 – I can’t seem to get a reverse lookup on it – no ptr record?

As always, proceed with caution when dealing with links in emails or files attached to emails.

sunbelt blog has yet MORE fake codec sites to watch out for. All are bad and should be AVOIDED… details after the jump….

IP: 85.255.118.195
vccodec(dot)com
IP: 69.50.188.109
hqcodec(dot)com
IP: 69.50.188.109
powercodec(dot)com
IP: 69.50.188.109
medcodec(dot)com
IP: 216.255.183.202
ptproject.com (currently offline)

These stories come up from time to time. A free giveaway of some sort and it turns out that there’s spyware or a virus embedded, company gives a big “whoops” and fixes things by replacing them…. McDonalds had a promotion going where up to 10,000 people could win a flash based mp3 player they also received a trojan horse preinstalled…. They’ve apologized and are swapping the infected players and giving information on how to clean up a pc with the keylogger. According to f-secure it was infected with the QQPass password-stealing trojan. Just imagine how things would have turned out if the Greeks had looked that gift horse from the trojans in the mouth first…..

Looks like Apples at it too – some of the 5.5gb ipods shipped with Windows viruses. ( RavMonE.exe ) – Apple has removal instructions….

Sans had an interesting malware analysis this morning about a blob that appeared to be ascii text (gibberish) that was retrieved by a piece of malware. It turns out that the ascii text was a cleverly encoded exe file (windows executable or program file.) It took several iterations of their analysis to uncover the actual file. A followup referred to a study of “hiding” malware in various Microsoft Word supported formats and how successful (or unfortunately UNsuccessful) several antivirus programs tested were able to identify it. This was performed by running the files through virus total and the virus was the EICAR test pattern.

The pdf of this analysis is available here. Basically, only 2 Antivirus programs were able to detect the malware in all 7 file formats. (They used .doc .dot .doc (rtf) .xml .mht .rtf .zip (word document, document template, Word 97-200 and 6/95 rtf, xml, single web page, rich text format and web page (which bundles all files into a zip archive.)))

It looks as though Microsoft and McAfee antivirus products were the only ones to raise an alarm over the file encoded in all 7 formats. Norton and Clamav were members of the 2 of 7 club and AVG (common free antivirus) detected 0 of 7.

So, maybe this should throw down the gauntlet to antivirus scanners to be able to better parse the contents of various files.

–Update 8/30/06–

Sunbelt blog has skewered the methodology of the above testing. Noting that the EICAR test signature is just a string of characters and nothing more. They even say antivirus programs SHOULDN’T detect this embedded in another file. They do say that this kind of testing is not entirely worthless, but should be done on real, live malware samples. (They note – that’s what VMWare is for…)

Also, on a somewhat related note, the security fix has a few musings on antivirus in wake of a scandal involving consumer reports inventing viruses to test antivirus scanners. in a way, the thinking makes sense – “I can’t test against KNOWN viruses, I need to test against new, unknown viruses…” Unfortunately, genies don’t always stay in the bottle, so that’s not a great idea for a number of other reasons.

He does not an idea that may have some promise – An idea that was presented at Defcon for a huge malware repository. The idea being that different AV vendors could submit their recent samples there, this would give others a chance to test their scanners against the new/emerging viruses. I don’t know how likely that is to happen, but it would seem to have some serious advantages. The biggest argument against is that it might make it easier for the “bad guys” to get at live working viruses, but realistically – they already have pretty easy access – maybe it’s time to give the “good guys” such easy access so that the playing field is level.

–Update 9/5/06–

After the criticisms of the original “hiding malware inside document files” test which involved the EICAR test signature, Jan Monsch has redone his tests. SunbeltBlog has a summary and the pdf. Really, the results aren’t that much different from the original findings with the test signature. AVG bats 0 for 7 and Microsoft and McAfee do very well at 7 of 7 detected. This time the test was taken with the Netsky virus. Clamav scanning for a postfix mailserver only cauight 2 of 7 of the netsky virus files. So the conclusion is that viruses aren’t usually embedded within a traditional document format in this sense. Could we see more of that in the future? If OpenDocument became THE “one format to rule them all” would it be a common vector? I think most virus writers will tend to follow the path of least resistance to the greatest benefit. In other words if it’s easy to write a networkable worm that will infect LOT’s of PC’s, they’ll do it. If that means embedding a virus within a document and embed the name as something that people will reliably click on, then that will likely work too. There will always be some that will target more niche areas, usually with a personal or monetary motivation. It’s worth it to know what vectors are possible and it would be good if all of the antivirus programs could do a good inspection of any mail attachments contents.

The metasploit project is now hosting a malware search that uses Google. It essentially uses a binary google search technique that was referenced last week to find malicious files hosted on the web. Of course, this will be partly limited by Google’s indexing which recently has not been quite as thorough as before, but… all you have to do is search by a virus name and find matches. I can see where this is useful for research. What I DON’T understand is why Google doesn’t integrate scanning of content into the googlebot indexing. It would take a lot of processor power. Well…. I think Google would come close to having enough to take a stab at this. I think they should AT LEAST…

Apply this kind of scanning to their paid advertisor links. I recall a recent article that said, not only was Google search (and other search engines) dangerous for finding malware, but the paid ads weren’t safe either.

In fact, I see the paid ads as a good proving ground for this type of google effort to make the web just a bit safer for browsers.

Hot on the heels of the Microsoft Word patch there’s a new threat to Microsoft Office. This vulnerability is with Excel documents. According to the MS security response center blog, they’ve received one report of a system being attacked by a previously unknown vulnerability in Excel. The moral of the story is to be suspicious of any attachments, be they programs, or claim to be images, word documents, excel documents, fluffy bunnies or what…. If you weren’t expecting an attachment in email wait and find out if it’s legit through other channels.

What’s really mind boggling about this is HOW MANY vulnerabilities there are that are known about by a select few and those select few have been “trafficking” in the market of undisclosed vulnerabilities or just taking advantage of them for their own use. I remember last fall there was an ebay auction for a claimed undisclosed Excel vulnerability. The reality is that it’s likely there are MANY vulnerabilities that have been found and never reported to Microsoft or publicly announced. (Other vendors as well.)

The best defence is to be suspicious of attachments and to always be cautious.

The last couple days, there’s been a virus spreading making use of yahoo mail’s interface. Usually web mail is considered a fairly safe way to get email, but in this case all that was done was the user clicking on a malicious email and the virus ran. It appears that javascript/AJAX/Web 2.0 applications are going to have to get closer scrutiny. In the Sans diary, they mention that they’ve analyzed javascript from several web applications and there are some that are vulnerable. (They’re contacting vendors.) They also point out web designers should keep this in mind as well..

The current worm could be readily modified to spread across many systems that do not escape javascript when displaying data from a foreign source. Many web developers should reexamine their code, and make sure that display functions do not deliver potentially malicious code.

The writeup on the yahoo worm gives the following details..

A Yahoo! mass-mailer is currently making the rounds with a subject of “[random word] New Graphic site”.

There is a good deal more that can be found at this link. Of course, turning of javascript kind of defeats the purpose of the mail interface. So that’s not a good workaround. They are working on a fix (already in the beta version) and are blocking many of the messages at this point.

Wolves in sheeps clothing are the label I give to those rogue antispyware, or antivirus programs that bring pests instead of protect against them, or are otherwise questionable in their tactics. Titan Shield seems to be a new threat on the block in this area, I haven’t seen it first hand yet, but it looks like it is one you’ll want to avoid *(You may want to block antispywarebox(dot)com and titanshield(dot)com if you’re in a position to do such things in your network.)

It’s almost surprising that this kind of thing is still a problem, but it seems that they just keep sprouting up. Anytime you’re considering getting antivirus or antispyware software, DON’T do it because of a popup window, investigate (offline if necessary) some of the legitimate, recommended options. Maybe check it against the list online of rogue antivirus and antispyware and then come to a decision.

Oh, also from Sunbelt blog… they’re talking about a craigslist car ad – on inquiring – the advertisor gave a link purportedly to details/pictures, etc. about a car for sale, the link was to an .exe file. You’ve got to stay vigilent.

It went un-noticed by most people for a few years. After all, the ones that were affected were just those that were “asking for it”. Where to start. Let’s see, back in the day there were some that sent out messages to other peoples computers and even when people tried to stop getting the messages they kept coming, so a few sites decided that if they could “blacklist” the places that these messages were coming from, they could help people deal with the mass of messages. So they did, and the people sending the unwanted messages were a bit frustrated and improved their distribution a bit, taking over virus infected pcs for sending their messages. The defenders matched and started blacklisting dialup addresses as mail sources. It was frustrating for those doing legitimate mail servers on a dynamic internet address, but there were legitimate ways to fix the problem. But the senders of the messages got mad.

They used the army of spam bots that they had cultivated to attack the blacklisters, from time to time they would deny service to their websites and frustrate the effectiveness of their service. Once, to show the power of a botnet several big name websites were taken down for a better part of a day, then the attack just went away. It was just a flexing of muscle. Of course, THAT made big news for a day or so, then life went back to normal. The attack against spam blacklisters continued off and on, but most people don’t really care about that. OH they hate junkmail, but they just don’t know if blacklisting is the answer. Maybe it’s not anyway.

Occasionally, the botnets were used for other kinds of attacks too. Not against anybody that you’d jump to defend. Mostly against gambling sites. They’d basically say, ok it’s going to take $$$$ to make sure people can still reach your site tomorrow. After all gambling is at the edge of the law in many places, they don’t have much wiggle room to contact authorities. So, again, people didn’t take big note of it.

Then, there was a company that had another idea for getting people out of junk message mailing lists. They would follow the law, which allowed a removal request to be sent for each message received. They had a download client that would automate the process, but stay within the law. The effect was close to a denial of service for some of the big junk mailers. Some quickly conceded and cleaned their lists, but some took the lists and turned on the users of the service, and ultimately there was a massive botnet attack against that service that went on and on, eventually causing them to close up their doors.

Then there was another site. They were dedicated to computer security, to helping people remove viruses and spyware and had started an initiative to take down phishing sites. They were a bit more “mainstream” I suppose than many of the other sites. They got noticed for their work and have fallen under attack.

The above is basically a true (although stylized) narrative of the last few years of online botnet activity. Now, I CAN’T CLAIM that these attacks were all made by the same group, certainly not. BUT, I think this list shows how powerful botnets have become and the threat that they pose to the internet at the moment. Castlecops.com is the site that is the most recent target of a denial of service attack. They seem to be up at the moment, but I am really beginning to think that the internet security community has a BIG problem and a BIG fight on their hands. I think the “take down” of blue security may have given extra confidence to many in the spam/virus/spyware/phishing “community” that they have the upper hand and I ask myself if we might see security related business and communities (like castlecops) targetted one at a time until they’re DOS’ed into submission.

Indeed, blue security talked about the next stage in their fight would have been an escalation and perhaps starting a full scale “war” on the net. So, the question is… how much does our economy depend on the internet? How much power then does a botnet yield that could take out major sites for a period of time? What solutions are there?

Most efforts at taking out botnets have gone after the IRC servers that act as “command and control”. Usually, blocking those is what’s called for. However, I am beginning to wonder if another approach would be better. I’m wondering, given the fact that if you have one trojan on your pc you likely have several…. if it wouldn’t be better to design a “white hat” upgrade to distribute to a trojan, so that on the next connect to the IRC control, it updates and then displays a “YOU NEED TO REMOVE VIRUSES FROM YOUR PC” message and disables all network interfaces (routes everything to 127.0.0.1)

I know many times such “white hat” viruses backfire, but I think there needs to be serious consideration of ways to take out entire legions of botnets at one stroke, rather than cutting off one head and then the bot downloads fresh code from another server.