Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
Security Spam The Internet Worms

Malware vs. Anti-Malware, 20 Years Into The Fray 62

jcatcw writes "Steven J. Vaughan-Nichols considers the dissimilarities between malware of yore and current infiltrations as we approach the 20th anniversary of the Robert Morris worm. Modern malware apps curl up and make themselves at home in your system, where they wait for a chance to snatch an important password or a credit card number. Welcome to the era of capitalist hacking. Any self-respecting malware program today is polymorphic, making signature-based antivirus approaches difficult. Heuristics and virtual sandboxes offer alternatives, but all such methods are reactive. Unfortunately, monitoring lists and networks is about the only current alternative."
This discussion has been archived. No new comments can be posted.

Malware vs. Anti-Malware, 20 Years Into The Fray

Comments Filter:
  • by Toreo asesino ( 951231 ) on Tuesday May 06, 2008 @11:38AM (#23313488) Journal
    Some malware i've seen has become seriously soffisticated, so much so cleaning it is basically impossible.

    Non-admin rights, client-side file-scanners, web-side black-lists, and user training is the only way malware is going to go away.
    • by wizardforce ( 1005805 ) on Tuesday May 06, 2008 @11:46AM (#23313596) Journal
      not only that but a more varied OS/software environment would lessen the damage that could be caused by malware/baddies in general. homogenization is likely one of the worst things to have ever happened to software in general.
      • more varied OS/software environment would lessen the damage

        Only if we don't unify our code, which probably won't happen because people will want to target broad user bases. When code can be compiled on a Windows machine to target a Linux machine, you still have problems.
        • Re: (Score:3, Funny)

          Surely then youd need to bundle GCC with your virus because most people dont have a compiler, meaning all viruses would be GPLd and you have to have to offer every machine you infect the source...hmm

          either that or you have to develop a self compiling virus, which has the chance of suffering random code changes and evolving...hmmm
          • OFF BUT RELATED TOPIC: I just wiped my Compaq laptop using the Manufacturer-supplied CDs. I barely had the "new" install turned-on 5 minutes, and suddenly I get a popup telling me to go visit registrycleanerxp.com (known malware).

            Is it possible Compaq sold me infected CDs???

            I shouldn't have a virus after a brand-new install.

            • by jotok ( 728554 )
              Was it online? SANS used to track the time-to-infection for an unprotected system connection to the internet (it was between 10 and 15 minutes a few years ago) and noted that this was less than the time it took to download and install a firewall, updates, patches, etc.
      • I disagree, but I have heard this a lot. I really have no idea where this idea comes from. If you have a wide variety of operating systems deployed, you are vulnerable to every new exploit that becomes available. Once on your network, almost any software deployment system is vulnerable to local network attacks, and then the rest of your network gets owned.

        I would argue for putting every OS deployment on its own vlan, and then using NAC to make sure they are all properly patched before getting out.

        Homogen
      • When you have a diverse collection of applications and ideas behind how those applications work, you have a continual flow of ideas. If, on the other hand, you have a monolithic OS that enforces "this is how you shall behave", you see less innovation and fewer new ideas.

        DUH.

        Diversity is HEALTHY.

        As opposed, for example, to forced quota-based mixing-up. As in college "affirmative action", which serves to homogenize colleges throughout the U.S. based on, of all things, "national averages" rather than
    • by GlL ( 618007 )
      I work for an ISP and my user base is a range from large corporations to grandma. I can't enforce 3 of your 4 suggestions, and the fourth could get me sued for violating someone's first amendment rights if I black-list the wrong IP range.

      Non-admin rights are fine to a point. There still can be compromise issues without admin rights. You can still compromise the administrators as well.

      Client-side file scanners are and will always be one step behind the bad guys.

      Web-side blacklists, while being the best way t
      • Minor point: unless your ISP is run by or funded by the the government (federal, state, or local), you would win a first amendment suit. The first amendment controls what congress can do, and is applied to the states by way of the fourteenth amendment. That is not to say that you wouldn't be hit by problems with common carrier laws, but the constitution has very little, if anything, to do with it.
        • by GlL ( 618007 )
          Actually, we are tied in with a local government who put in fiber and cable, and maintains the routing, so that adds "fun" legal complexities to our mix.
    • Non-admin rights, client-side file-scanners, web-side black-lists, and user training is the only way malware is going to go away.

      Yeah, 'cause we've seen how great all these methods have worked so far. We're using them for 20 years now and malware's doing better than ever.

      You know what I find interesing about all these methods you listed? They all assume that security has already been breached, that malware is on your computer, and attempt to contain damage and patch things up.

      Is it just me that finds this

    • Non-admin rights, client-side file-scanners, web-side black-lists, and user training is the only way malware is going to go away.

      I wish I had written the Checklist Form For Why Your Anti-Malware Idea Will Not Work already.

      Non-admin rights : yeah, right. "Your mouse has moved : cancel or allow"?

      Client-side file-scanners : reactive security. Useless. Moreover, there are tons of ways to hide malware so that file scanners don't see them.

      Web-side blacklists : Not going to happen. When the file is on the blacklis

  • Robert Morris, OMG (Score:5, Informative)

    by Anonymous Coward on Tuesday May 06, 2008 @11:43AM (#23313560)
    Come on, the guy's name is Robert Morris:

    http://pdos.csail.mit.edu/~rtm/

    You're thinking of the William Morris talent agency in Hollywood, or something. Mods, please correct this.
  • Some ways to win. (Score:3, Informative)

    by apathy maybe ( 922212 ) on Tuesday May 06, 2008 @11:49AM (#23313622) Homepage Journal
    Don't install system wide untrusted software, only use signed software from your public repository or from trusted vendors.

    Prevent any other changes from being made to the system, mount system partitions read only.

    Where users are installing software, force it into a sandbox (one for each application). Each sandbox will have limited access to the network, user files and hardware (such as web cams and microphones).

    The simplest solution is to never allow software from users to run (mount home partition as no-exec). However, this doesn't cut it much of the time, which is why I would suggest doing something similar to no-exec, but as a sandbox rather then not running the file at all. I'm not sure how hard that would be, but I'm sure it is possible.

    (Oh wait, are we talking about MS Windows here? I guess you can ignore what I said then...)
    • only use signed software from your public repository or from trusted vendors.
      How does a vendor become trusted under your best practices?
      • A vendor is someone with a name, a face, a business name, an email, a phone number, a building address and so on. A trusted vendor depends on your criteria and could range from the length of time a vendor has been in operation (i.e. don't trust a two week old start up), the size of the company (a two person job might be less trust worthy then a 100 person place), and other such possibilities.

        Of course, when it comes down to it, do I trust Microsoft? Well, no, they have a history of making buggy products, ev
        • by tepples ( 727027 )

          A vendor is someone with a name, a face, a business name, an email, a phone number, a building address and so on. A trusted vendor depends on your criteria and could range from the length of time a vendor has been in operation (i.e. don't trust a two week old start up), the size of the company (a two person job might be less trust worthy then a 100 person place), and other such possibilities.

          For PCs in a home environment, would you recommend criteria that shut out software developed and self-published by a micro-ISV [wikipedia.org]?

          I don't know of any tool similar to System Restore Points from MS Windows in GNU/Linux land, but dd could easily do a similar job.

          Especially considering the more robust separation of programs and read-write data that the *n?x mindset has always encouraged.

          • It isn't my job to decide what is a trusted vendor for you. That's your job.

            Obviously single developer outfits are going to have more trouble being "trusted" if for no other reason then their signing key is not signed by some key company.

            Personally, I tend to be a lot more willing to download random things off the web if:
            The software is free software.
            The website "smells" clean.
            I've heard of the software (or had it recommended to me).

            But that's for X/GNU/Linux, what about MS Windows? Well, I'm a lot more war
  • by nuzak ( 959558 ) on Tuesday May 06, 2008 @11:58AM (#23313736) Journal
    Wish I could get paid just for clicking "approve" and filling in the text in the "from the ____ dept".

  • by mcelrath ( 8027 ) on Tuesday May 06, 2008 @12:10PM (#23313896) Homepage

    Between spam, malware, and credit card fraud, the criminals are winning, big time.

    The eventual consequence of this is a faltering of trust in our financial systems and economies, and the rise of new kinds of criminal mafias, with billion dollar portfolios. If you thought the mob was scary, wait until you see what rises out of the ashes of the current system.

    The solution to this, I believe, is first to limit the information transferred in any transaction to that which is necessary for the transaction (no grocer, you don't need to know where I live); second to implement electronic cash (in the current credit card system you give authorization to perform transactions at any time in the future without verification); and third to establish and teach strong cryptography for communications, transactions, and identity.

    But the biggest thing we can do now is get the world's police forces to get off their asses. As long as these things are not prosecuted, criminals will flourish, and they are.

    It's time to make this an important issue in elections, before we all lose big.

    • But the biggest thing we can do now is get the world's police forces to get off their asses. As long as these things are not prosecuted, criminals will flourish, and they are. Word. The behavior you reward is the behavior you will get. The current non-system rewards malware with little downside. Really, voters don't like malware, why hasn't some ambitious commonwealth's attorney or state attorney general gone after this?
      • Really, voters don't like malware, why hasn't some ambitious commonwealth's attorney or state attorney general gone after this?
        One did, but then they made him governor and he got caught banging a hooker or something.
    • radio shack was the worst about this. you'd tell em that you were buying a dollars worth of random substandard parts for a quick project and they wanted your address and phone number.

      tell em no thanks and they would chime in that they could then send you catalogues. well what was my phone number for?

      i found it easiest to tell em my name was "jim john joe billybob" and my address was 123 A St. and make up a phone number with the local area code. then complain if they thought it was hokey. i wonder if jim
  • by Anonymous Coward on Tuesday May 06, 2008 @12:11PM (#23313906)
    The whole way security is treated needs to be changed. Having root and an ordinary user just doesn't offer the level of granularity that users need. As a user I want to be able to do everything on my computer, what's really needed is fine grained access control per program. Of course, that has issues with users having to grant those privileges but you could have profiles. Imagine installing Evolution or something and it pops up and says "This software says it's a mail client, does that sound right to you?" and then what privileges it gets granted will be set by a "mail client" profile already installed on the system.

    When you need to install something esoteric then you would have to do some more advanced steps but if you are installing something strange then you probably know what you are doing anyway.

    This could maybe be combined with some sort of trust network. Say your friend installs something that needs non-standard access rights, they could grant the required permissions and create a new profile. You would have them in your trusted list and would have access to all of their profiles so when you install that application, it can categorise it using the info your friend provided.

    I think this system provides a good balance between really fine grained permissions and not blindly clicking through loads of confirmation dialogs.
    • That sounds good, but I don't think the average user has the patience for a system like that. I know even as an IT guy it would get a little old. And unfortunately I don't trust the software venders to come up with an effective automated system that would lessen the required user intervention.
    • even then Malware can still read your user data files and upload them to the bad guys.
    • by drsmithy ( 35869 )

      The whole way security is treated needs to be changed. Having root and an ordinary user just doesn't offer the level of granularity that users need. As a user I want to be able to do everything on my computer, what's really needed is fine grained access control per program. Of course, that has issues with users having to grant those privileges but you could have profiles. Imagine installing Evolution or something and it pops up and says "This software says it's a mail client, does that sound right to you?"

      • The way I imagined program access control was that each user program would be confined to a sandbox where it only had access to its application directory, a (system configured) data directory for that application and basic OS API functionality. If an application wants network access, it can request it from the OS and the OS will ask the user whether to grant the application the access it seeks. The user could be informed by the system what the implications of granting access would be. It could prompt a sing
  • I just don't understand why malware isn't considered a form of vandalism and prosecuted as such.
    • I don't think it would work. The same way the war on drugs doesn't work and never will. When people want/need something bad, they'll pay for it. When there is enough money involved someone will use all the means necessary to provide those things. Making anything illegal just creates more criminals and sociopaths. That in turn leads to more police which are anyway fallible to corruption. Don't fight the effects, fix the source of the problem. Make people aware of the dangers of malware. I know it's been said
      • TO be fair, distribution/creation of malware -is- something that can be stomped upon. Various juristictions have computer use/abuse/misuse legislation.

        But either way, it's like trying to change the weather by pissing in the wind. Isn't going to do much, apart from getting you dirty.

        The _only_ solution to people using these powerful, complicated tools, without making a complete mess of them, is by adequate knowledge and training.

        Until that exists, NOTHING that you do, precaution wise, is going to do a

  • "Welcome to the era of capitalist hacking."

    What does the theft of personal information have to do with the private ownership and exchange of wealth?
    • What does the theft of personal information have to do with the private ownership and exchange of wealth?

      1) Hack system (manually or via malware)
      2) Steal personal information
      3) Sell personal information
      4) ?????
      5) Profit

      Does "Hacker for hire" or "For profit hacking" work better for you? The correct term is "Cracker". Either way, it is a capitalist system that they function under. There is demand for a good (personal information). The cracker answers the demand by producing a supply of that good.

  • by MadMidnightBomber ( 894759 ) on Tuesday May 06, 2008 @12:42PM (#23314316)
    Everyone knows it was Philip Morris, the guy who makes the cigarettes.
  • by Arrogant-Bastard ( 141720 ) on Tuesday May 06, 2008 @01:13PM (#23314738)
    In re: "Unfortunately, monitoring lists and networks is about the only current alternative."

    There are many alternatives to this, starting with: "Recognize that operating systems which are readily compromised by malware are broken and not acceptable for use." If you choose to use an OS which is so intrinsically weak that it cannot survive exposure to the (unfirewalled) Internet without anti-virus, anti-spyware, anti-adware, etc., then you have chosen poorly, and no subsequent choice you make will compensate for that.

    A followup point would be "Understand that it is not possible to 'clean' a malware-contaminated system. The only acceptable course of action is to wipe to bare metal, reinstall, and restore from backups." While it might have been partially true in a limited sense that some malware could be removed by anti-whatever products, that's certainly not the case now: it's much more likely that malware will evade detection and removal. Of course, it serves the purposes of both anti-whatever companies and lazy system administrators to continue propagating this fiction, because if they actually had to scrub and rebuild systems as often as they're infested, they might have to face some hard choices that they'd rather not.

    And an excellent set of auxiliary points may be found in Marcus Ranum's The Six Dumbest Ideas in Computer Security [ranum.com], where he enumerates the most egregious (and sadly, most common) mistakes made by nearly everyone, including supposed "experts" with strings of meaningless, worthless certifications after their names.

    So there are plenty of alternatives -- but choosing them and implementing them requires vision and insight, two qualities badly lacking in many in the profession.

  • Alarmist (Score:4, Insightful)

    by redelm ( 54142 ) on Tuesday May 06, 2008 @01:30PM (#23314978) Homepage
    Sorry, I'm not paranoid. Go peddle your fear somewhere else. Yes, there are real threats. There is also a cost both in money and peace-of-mind of fighting them.

    There is a balance to be struck, and "Better safe than sorry" can be answered "better neither than either".

  • One page print page [computerworld.com].
  • the "William Morris worm" sends you scripts, tries hard to get you to take a meeting, then charges 15%.
  • Working with residential users on a regular basis, I have come to repeat: "There is no One Program that rule all malware (and I explain malware includes all the crapware since They figured out profit was to be made online). Safe surfing habits are the best defense against Malware." Soapbox aside, since Sony released rootkits into the wild, I have had more success with backing up data and performing the elegant Nuke and Pave. Format and reinstall. Without doing that, I have very little confidence in any
  • Why have we seen no 'terrorist malware'?

    I would naively assume that it would be easy enough to buy off the shelf botnet code release it and when it gets to a sufficient size upload something really toxic. For bonus points the attack could be limited via IP address or targeted at idealogically unsound files.

    From a practical POV this sort of attack would circumvent the normal surveillance as there is no need to go to terrorist camps, no need to buy suspicious chemicals ... they would still need to keep th

    • I think it's more that a distributed denial of service attack, is just plain unexciting on the grand scheme of flying aircraft into office buildings.

      No one ever died as a result of a computer virus.

      • by mlush ( 620447 )

        I think it's more that a distributed denial of service attack, is just plain unexciting on the grand scheme of flying aircraft into office buildings.

        No one ever died as a result of a computer virus.

        I wasn't really thinking of DoS ... how about on 8th August every infected computer overwrites its hard disks with copy's the the Lampton manifesto.

        There are bound to be a few infected computer in hospitals airports, power stations etc. and it does not matter if they were not attached to anything important, the news story's will be all about how the Lampton worm nearly caused planes to fall, patents to die and 'endangered' the Grimbledown nuclear power plant.

        Later on they will move to human interest

  • I am reminded of a rather amusing sci-fi short story I read. I think it was Cory Doctorow's 'Robby the Row boat'.

    But anyway, one of the ideas it espoused was that malware is what's driving systems development to the point of passing Turing tests. Between captchas, baysian filters, and similar 'proove you're a human' malware countermeasures, with virus heuristics, and malign software detection, you have a very potent 'reaper' process, which kills off substandard malicious code.

    The stuff that sticks, is the

  • Instead of reactive solutions, better computer architecture could be a solution.

    A so-called "worm" always spreads by injecting and executing its code into a vulnerable process on a remote computer. For example, on an IBM AS/400 it can not do this, because if you overwrite a pointer with data, then it is not a pointer anymore - so it can not be used to address memory (that's why the machine actually has 65 bits instead of 64 bits, the 65th bit is a tag flag that marks pointers. aka pointer in memory protecti

Keep up the good work! But please don't ask me to help.

Working...