'Bagle' Worm Heading For A Windows PC Near You
mrSinclair writes "the 'Bagle' or 'Beagle' worm is expected to hit the U.S. by midweek, probably Tuesday as many employees return from a three-day weekend." He points to this Washington Post story (via Yahoo!), which describes the Windows mass-mailing worm as being transmitted via email as an .exe attachment and as installing "a program that lets attackers connect to infected machines, install malicious software or steal files." The article says Bagle has been detected in more than 100 countries. Other readers have sent in links to coverage at the BBC and at SearchSecurity.com.
Why is this one unique? It's just the next worm.
And it replicates by *emailing* itself...
No remote root/admin exploits, no network-clogging mass scanning, no nothing.
Maybe just a few malconfigured mailservers going down, that's it.
yawn, wake me up when we're at threatcom 4
First, you'd have to save it to your hard drive, clicking on it wouldn't work (email attachments are data files, not executables). Then you'd need to "chmod +x" it, and then you could run it as your user, in which case it can infect only things associated with that user. Assuming these unlikely things happened, the superuser can simply disable your account and clean things up, while everyone else on the system can chug along happily.
In other words, its not the same. Unix made the right decision from the beginning to separate data and executables, and to keep most users at a non-Administrator/non-root capability level.
> Then you'd need to "chmod +x
.py, etc) and just go launch the script intepreter when you double-click on the file. This does not require +x access!
This all really depends on how much "Shell Integration" your Unix desktop has.
It's quite possible that a Unix Mailer would look at the file extention (.pl,
KMail was caught launching PE EXE viruses using Wine for example.
In reality, most of these mail viruses have nothing to do with OS security and everything to do with poorly designed mailers and dumb users.
Right. Mass migration to FreeBSD, Linux, Mac OS X. Massive porting of all possible windows apps to Unix. Suppose that whould happen quickly or even overnight. You can always hope.
Will the problem become less severe? Probably, at least for a while. Will the problem go away? Of course not.
Because insecurity stems not from some flaw in an OS but from a fundamental problem with the users and industry's mindset which stresses features and convenience over security. Just imagine what a simple script could do on a Uix dervative when accidentatlly run aby a user. Now imagine what happens when that user is running as root. And that's just what many people are going to do...
Once in a while, I even pass the Turing-Test
I do this as well. .exe files can be sent through these by renaming the file (e.g. to .jpg), then adding a comment "please rename the file to .exe".
.jpg or .gif, and with the added guidance for the receiver. Of course it was again blocked by my scanner, but apparently this method works on the commercial scanners and the users know the workaround.
.zip and telling the user to unzip and then run the program.
Of course you must make sure you use a valid detection mechanism.
Many commercial scanners use the extremely naive approach of checking the file extension!
This means that
You would not believe it, but even the most well reknowned scanners use this stupid method. I have seen countless examples of "funny programs" being blocked on the mailscanner, and then the same file arriving half an hour later, renamed to
There even has been one trojan that uses this method by packing the program in a
Two main reasons - the extra load generated and the risk of false positives.
If filtering were done as you suggest, with a simple attatchment file size check, then there's a reasonable chance a perfectly legitimate mail would be dropped. It also wouldn't take very long for the virus writers to create viruses that vary the file size on every reproduction.
If a customer gets themself infected with a virus then it's their fault for not have adequate virus protection - if the ISP drops their mail because it was of a similar size to a virus it's the ISP's fault.
Blaming GW Bush for the Iraq war is like blaming Ronald McDonald for the poor quality of food.
We had the same executable attachment problem back when I was in school in the late '80s. Our VM Mainframe E-Mail system got shut down because of some christmas card program that remailed itself to everyone in your address book. Sound familiar?
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
I don't know whether it applies to that one, but a _very_ efficient way to avoid the annoyance of Windows email worms is to use your firewall block all incoming traffic from a Windows machine to port 25.
:
On OpenBSD, the following line is enough
block drop in log quick proto tcp from any os Windows to any port smtp
There is really not a lot of legacy mail exchangers running Windows so it doesn't hurt.
However, it blocks most worms that are trying to directly send mail.
{{.sig}}
Je fume. Tu fumes. Nous fûmes!
The virus exploits the massive Windows bug that clicking on an attachment is enough to run an executable with full user priveleges (root privileges, often) and that there is no safe mechanism to _open_ a file without the risk of _running_ it.
-- Ed Avis ed@membled.com
I don't see how pipes are such a nightmare. It makes sense to allow programs to direct their input and output to eachother without needing to use an intermediate file. (And WinNT and its derivatives have pipes as well, so it's not like it's a UNIX-specific weakness.)
RPM hell is pretty much gone in any mainline distribution these days, what with apt-get, yum, emerge, urpmi, and yast's online updating. All of the major distributions have a free way for you to update your system with full dependency checking and resolution. Even Slackware's got it with swaret.
If you don't think KDevelop is a "real" IDE you might want to look again. The newest release, based on the Gideon codebase, is astounding. Code completion is only part of the good stuff included.
OpenOffice is just about the same as MS Office - I haven't seen any compelling reasons to use Microsoft's version instead, especially considering that OpenOffice runs on my OS and MS Office doesn't (at least, not natively).
The technology is pretty much in place at this point. There might still be a few straggling areas (games are a sore point at the moment, but more and more developers are releasing Linux versions these days than ever before) but on the whole, Linux on the desktop is just building momentum, and nothing is stopping it. It'll hit critical mass sooner or later, and once it does, it's game over for Microsoft. I don't really care personally when it does for the rest of the world - I'm happy with it right now.
Anyway. Good times. Use what works, as that's what you need. But you might be surprised if you try out a mainstream distro, as a lot more works these days than ever has before. And no, FreeBSD isn't even close to mainstream. I love FreeBSD5 and I'm using it (with pf) on my firewall, but I use Linux on my workstation.
You could create a priveledged system since NT. Heres a scenario for you, Linux comes preinstalled on every new computer sold and is the dominant OS. Do you think resellers would setup non-root/non-rootlike accounts for the user? It's not like they couldn't do that with 2k or XP. And what about the bagillion possible daemons that the reseller might turn on just to make things even easier for the user? do you think the reseller would educate the buyer on the importance of actually maintining a system or firewalls? *nix (as much as I love it) is not the be all, end all to this little annoyance. Education is. If people were educated on how to actually use their machine, this problem wouldn't exist.
At least with Windows Update, the user can be assured that they will get a secure untrojaned binary. No one has any evidence that Windows Update has been rooted.
Of course six months from now, when they finally get around to issuing a patch, the lack of source code also leaves no evidence that a new vulneralibility wasn't created when the old one is closed, does it?
Time is what keeps everything from happening all at once.
Well in any case it should be a non-issue. If you are running Windows correctly, you're not running as a member of Administrators but rather a regular user with all the permissions correctly set. This way you can't inadvertently destroy data that should be secured (e.g., programs). In any case, I have grown tired of attempts to trivialize the would-be damage of worms on UNIX systems as "oh it will only trash /home/user" -- as if that's not bad or something!
(Also of note is that most people sending these worms unbeknownst to them are home users, not corporate users on multiuser systems.)
Except half the Windows programs out there refuse to run as a regular user, as they expect to have write access to system level directories. Consequently it is generally not practical to run Windows as a regular user.
So basically it exploits user stupidity. Thanks for putting it so eloquently :)
Why? you can easily write a userspace smtp client for linux, which is what this virus is. add it to .bash_rc or similar and away you go, each time the user logs in they start hammering away with copies of itself. Then, after 2 weeks, have it wipe out every file it can on the system - sure the OS will survive but plenty of what the user considers vital information will be lost.
Backups are just as required in Linux as they are in Windows.
Ewan
Yeah, but how much time do you spend trying to make sure you don't get anything? Searching for viruses on my 2.8GHz SATA 150 through less than 30GB of data on a RAID 0 drive takes HOURS. Then another 5-10 minutes everytime you install a program to make sure it's not kitted with spyware and such crap. Besides even normal users can install stuff in linux (contained to their home directory, only), whereas you cannot in windows, which forces Windows' users to Admin up EVERY time , which GREATLY increases the virus' accessibility. Plus the file structure is alot more accessable to normal users in Windows. Remember, the UNIX backbone has been around WAY before Gates stole DOS from that poor guy. If Windows users didn't have to admin up so much, they would be less inclined to log in as root all the time. I mean, even the "Run as.." function is hidden in windows! you have to hold the Shift key down while right mouse clicking to get it! If they can't figure out how to run as/su without jumping through hoops, of cource they are going to login and run everything as admin. I NEVER run Linux as root, I ALWAYS run windows as admin. It's just too much of a pain in the @ss in windows. Does the world need better PC education, or a better OS? I think we need both.
Of course you can do it in windows. But close to everyone in Windows runs as admin, while close to no one in Linux runs as root. In effect you wouldn't have to change anything in Linux, while you'd have to drop all your admin privileges in Windows. I've tried maintaining an XP box, and on most occations I had to log out my normal user and log in as admin because the run as feature simply did not work properly. Games couldn't run because the permissions were wrong, and impossible to change to the right ones (I tried, I called friends of mine who are windows admins, who told me it was different on their XP boxes...). ... I don't believe it's as easy to do to every windows box as it is to every Linux box. In Linux all you need is the capacity, in Windows the users need to refrain from using their default proile. Big difference!
------- I fumbled my registration and I now must suffer
Spamassassin is great...
However, people likely to get hit by this "bagle", is very much unlikely to be able to operate their own server running procmail + spamassassin.
If everyone repeats this refrain enough people may actually start to believe it, and that would be good in counteracting that old 'many eyes make all bugs shallow' phrase we keep hearing about open source.
Taken at face value the statement seems reasonable, but I'm a scientist and I like to hold theories up to the light of reality and see how they do. I know that testing theories annoys people because it makes them question their deepest held beliefs, but hey I'm an annoying guy anyway.
We could test the statement by finding an Open Source project that has much more market share than a closed source project, then compare the rates of exploit. Hmmmm... how about Apache vs. MS IIS?
According to Netcraft Apache has about 67% of the market and Microsoft's IIS has about 21% of the market. The often quoted FUD says that Apache is used by so many more people it must have many more exploits.
We can search the CERT website for the terms 'Apache' and 'Microsoft IIS' clicking on the boxes for :
Advisories
Incident Notes
Security Improvement Modules
Vulnerability Notes
'Apache' gives 180 results.
'Microsoft IIS' gives 830 results.
Wait! That means that just because something is used much more widely than another thing it does not result in more attacks! That proves the statement that if Linux were used more it would have more viruses is a false statement! It could be that open source actually does produce more secure code after all!
If Linux had 60% or 70% market share, there would probably be more viruses written for Linux than there are now. But, as we can see with the real world example of Apache and Microsoft IIS, the open source development model produces more secure software.
Sorry to step on that often quoted line about linux and viruses, but I like reality.
As a scientist, I'd think you'd know that only using one data point is not 'holding it up to the light'. I'm not saying the OP is correct, but you haven't proved anything, except that IIS has more reports on CERT than Apache does.
IMHO there is a delicate balance between security and getting the job done.
In many organizations, the developers are under the gun to meet project deadlines. You are more likely to get in trouble for not meeting a deadline than for running X as root.
Similarly, the system administrators are rated by how smoothly things run. Taking a chance by allowing developers to run things as root does not do them any good.
Sadly, from a developer's perspective, system administrators are rarely rewarded by their managment for helping developers sort out all the permissions issues.
If this is done, then one can figure how to set up the non-root account to get the work done without creating security problems.
It doesn't help that developers are often considered "knowing enough to be dangerous."
So system administration managers sometimes set the tone of "lock down the developers so they can't get away with anything."
One place I worked had the development servers locked down so tight, it was said you could only test in production.
Through my career, I have seen a lot of development move from the Unix platform to the Windows platform, partly for this reason:
1) The Unix System Administration department doesn't care about windows boxes, so they don't bother to control them.
2) The Development department knows that they can set up a bunch of windows boxes, give themselves administrator access.
3) The development project proceeds quickly in terms of accomplishing the project goals. The development manager is not rated on how few security holes he sets up in the process.
4) The managers learn: "Wow, if we bypass the Unix System Admins, we get projects done so much faster."
It is unfair to blank admins for security holes created by developers.
It is unfair to give an agressive deadline to the developement department and then ask them to work with a system administration department that has no incentive to help you meet your project deadline.
"We can't solve problems by using the same kind of thinking we used when we created them." -- Albert Einstein