eWEEK content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.
The rapid spread of MyDoom.A was a pretty scary thing to witness. I knew that I had immediately recognized it as a worm when it showed up here, but still its scary when so many people out there get infected. Then MyDoom.B came out and I really got concerned, partly because of its evil practice of locking users out of accessing security sites from which they might disinfect themselves.
MyDoom.B seemed like a slam-dunk to spread far and fast. Apart from the usual mass-mailing and KaZaA-based propagation methods, it also searches the Internet for systems with the MyDoom.A backdoor installed (and it uses a really weird method of scanning for those systems, skipping many of them for no apparent reason). When it finds a system running the backdoor, it sends it a copy of MyDoom.B for installation.
In the abstract, this should be the right way to do it, and youd think that with an ecosystem so fertile with MyDoom.A infestations, that B would be all over the place. Such is emphatically not the case. I searched the analyses of the MyDoom.B virus on the Web sites of several security firms. I found little reason for fear; the technical descriptions are all pretty scary, but almost all of those sites with an assessment of how far it has spread classify that spread as “little or none.” Heres a handy list for your own inspection:
-
Links to Security Firms MyDoom.B Analyses and Remarks on Spread in Wild:
- Symantec: rates it “Low (0-49);” separately, Symantec Security Response is seeing less than a dozen submission of the B variant
- Trend Micro: found 1 copy in the wild
- McAfee: “Low-Profiled”
- Kaspersky: No comment on distribution
- Sophos: “At the time of writing, Sophos has received no reports from users affected by this worm.”
- Panda Software: Distribution: Low
- Norman Antivirus: LOW RISK
- Command Software: No comment on distribution
- F-Secure: LEVEL 2 ALERT (“New virus causing large infections. Might be local to a specific region.”)
- TrueSecure: No comment on distribution
- Aladdin Knowledge Systems: Threat Level: Low
Interesting that even though there was a lot of disagreement on the name of the A variant (MyDoom.A, Novarg.A, MiMail.R, etc.) they all call the B variant MyDoom.B.
The only company that rates it anything above nothing is F-Secure, a company thats done a bit of scaremongering in the past. But “LEVEL 2” indicates less than it may seem, since the next lowest level of severity, LEVEL 3, describes a virus not necessarily in the wild. In other words, their levels and descriptions are not sufficiently granular.
Next page: So what will happen on Feb. 3?
So what will happen
Feb. 3?”>
And I did get one big caveat from an analyst I respect. Ken Dunham, director of malicious code for iDefense Inc., reminded me that the availability of the backdoor in both opens some possibilities that are difficult to predict. There could be some unknown capability being introduced through it, and perhaps we just still recognize the mutated infection as a simple MyDoom.A. Its possible, but its also been many days now that experts have had these viruses in their possession and they have been analyzing many honeypot systems. If there were something more going on, I think wed know about it by now. Perhaps theres some important technical flaw in the worm and the companies dont want to discuss it publicly.
So on Feb. 3, when the B virus launches its planned DDOS attack on www.sco.com and www.microsoft.com, will the world end? It appears that Microsoft has very little to fear. There arent enough infected systems out there to raise a stink, and theres no reason to believe there will be any disruption of service. As for www.sco.com, its already temporarily out to pasture from the MyDoom.A attack, and MyDoom.B wouldnt have changed things much.
The small number of systems with the infection made me consider the possibility tracing the worm back to the author. If there really are just a few out there, perhaps careful forensic examination of them can identify the initial seeding of the virus and even the source of the initial seeding. I asked Chris Belthoff, senior security analyst at anti-virus company Sophos, who said that—sadly—it doesnt work this way. There are thousands of lesser viruses that dont spread beyond a few systems, and the authors of these typically dont get caught. I guess youd have to make a stupid mistake.
As Netcraft observes in its analysis, Microsoft could be taking defensive measures as it has done in the past, such as giving Akamai or some other content distribution network the front-end duties for serving microsoft.com. The company has not done so. www.microsoft.com still points to a series of Windows servers on IP addresses owned by Microsoft. I guess Microsoft doesnt see all that big a threat either.
Isnt it interesting that these mass-distribution virus authors usually make some important technical error? The reason, in a very general sense, could be that its not easy to write quality software that can run on a wide variety of systems, even if they are all just different versions of Windows. This is one of the reasons companies have beta tests and hire independent testing firms. Alas, virus writers have a problem in arranging beta tests, and even to the extent that the A variant is a beta for the B and C variants, they dont get all the feedback they would want. Heres to non-standardization!
Security Center Editor Larry Seltzer has worked in and written about the computer industry since 1983. Be sure to check out
More from Larry Seltzer