Stories
Slash Boxes
Comments

News for nerds, stuff that matters

Slashdot Log In

Log In

Create Account  |  Retrieve Password

Securing IM and P2P Applications

Posted by samzenpus on Wed Dec 28, 2005 01:37 PM
from the share-safely dept.
Ben Rothke writes "Noted security veteran Bruce Schneier has observed that for those organizations that have incorrectly deployed cryptography, it is akin to putting a big flagpole in front of your facility and hoping that it will stop any attackers from breaking in. Of course, any attacker with intelligence will simply go around the flagpole rather than running into it." Read the rest of Ben's review.


Similarly, many organizations have deployed myriad security hardware and software products in their infrastructure. But when it comes to instant messaging and peer to peer applications, these applications often execute below the radar of many security products. This is due to the fact that the security infrastructure in many organizations was not architected to deal with such applications. These applications often have so much functionality that it obviates much of the security afforded by the security hardware and software products.

Using file transfer as an example, many organizations have policies and controls in place to stop the use of protocols such as ftp and tftp. This is fine, but that will only work for the ftp protocol. File transfer can still be carried out by most instant messaging clients, and that can pose serious security risks.

With that, Securing IM and P2P Applications for the Enterprise provides an excellent overview on how to handle, manage and secure IM, P2P, and IRC applications. This book is written for security and system administrators that need specific details on how to control and secure IM, P2P and IRC applications in their organization.

The need to get a handle on IM and P2P is crucial given that IM has turned into a global communications medium with most organizations today reported that they allow it for business usage. Many marketing and technical support calls are now handled via IM and this translates in to well over 250 million IM users worldwide. P2P is great for downloading music and movies, but that that poses serious security and legal liability risks when done on most corporate networks.

But with all the benefits that IM provides, it introduces many security and privacy risks. IM viruses, identity theft issues, phishing, spyware and SPIM (SPAM over IM) are just a few of the many risks. These risks can turn into intellectual property losses and legal liability issues especially when they are combined with targeted attacks on corporate IM users. Companies that don't have an effective way in which to deal with IM and P2P are in serious danger as most IM and P2P threats fly under the radar of many traditional security solutions.

The book has a fairly straightforward approach. Chapter 1 provides an introduction to IM and the most common security issues that IM brings into an organization. The bulk of the remainder of the book details various different IM applications in Part 1 (AIM, Yahoo, MSN, ICQ, Google, Skype), P2P applications in Part 2 (Gnutella, eDonkey/eMule, BitTorrent, FastTrack) and IRC networks and applications in Part 3.

Each chapter details the specific architecture of each application, its protocols, security issues, and solutions in which to secure the application. System administrators can use many of the checklists to quickly perform the initial steps necessary to secure their organization from unauthorized IM, P2P, and IRC applications.

Each chapter also provides significant details about the internals on how each application operates. In addition, various 3rd-party tools that can be used to secure and limit the various applications are listed.

Many companies are finding that a significant amount of their bandwidth is being used by P2P applications and Part 2 describes how to secure networks from the use of P2P applications. This is not always an easy thing to carry out given that many P2P applications, such as Gnutella are designed to easily bypass many of the security control mechanisms placed against it. Administrators will find that in this case, simply blocking Gnutella ports will not block all Gnutella traffic and the application still will be able to run. What is required in this case is the use of a firewall that supports deep packet inspection. Chapter 9 helpfully lists the commands to use when using iptables to block Gnutella traffic.

Chapter 12 provides an interesting look at FastTrack, which is the P2P protocol and network used by clients such as Grokster, Morpheus and other file sharing programs. The chapter also uses Ethereal to detail the internals of FastTrack.

Part 3 deals with IRC and is the sparsest part of the book. This is due to the fact the P2P and IM are much more heavily used on enterprise networks, which this book is geared to.

The only negatives about the book are its price, and some of its formatting. At $49.95, it is on the higher-end of computer security books, with the majority of such titles being in the $25.909 - $39.99 range. The formatting uses a font size that is somewhat larger than other book. This seemingly serves to achieve a high page count.

In addition, the book often references tables of secondary information that spans a few pages (for examples see pages 72-80, 115-120 and more). Such information would be better served in a multiple-column table in a smaller font. Printing the information in such a manner can cut down on the page total, and save a few trees at the same time.

Besides those two minor issues, Securing IM and P2P Applications for the Enterprise is a most helpful guide. Security and system administrators can use the book to get a handle on the increasing number of IM, P2P, and IRC applications that are found on the corporate networks they support.

Ben Rothke, CISSP is a New York City based senior security consultant with ThruPoint, Inc. and the author of Computer Security 20 Things Every Employee Should Know (McGraw-Hill 2006) and can be reached at ben@rothke.com"


You can purchase Securing IM and P2P Applications for the Enterprise from bn.com. Slashdot welcomes readers' book reviews -- to see your own review here, read the book review guidelines, then visit the submission page.
+ -
story
This discussion has been archived. No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
 Full
 Abbreviated
 Hidden
More
Loading... please wait.
  • Hey! (Score:5, Funny)

    by Phae (920315) on Wednesday December 28 2005, @01:43PM (#14353135)
    Of course, any attacker with intelligence will simply go around the flagpole rather than running into it.

    Hey! Are you calling me stupid?
    • Re:Hey! (Score:5, Funny)

      by ackthpt (218170) * on Wednesday December 28 2005, @01:49PM (#14353178) Homepage Journal
      Of course, any attacker with intelligence will simply go around the flagpole rather than running into it.

      Hey! Are you calling me stupid?

      I've been wondering about all those dents in the flagpole, about 5.5 feet above the ground...

      that ringing, an angel just got it's wings!
      No, wait, it's that guy running into the flagpole again...

  • Microsoft's doing their bit by including UPnP in new version of Messenger and encouraging people to use it.
      • Still, in order to use it, you have to TURN IT ON at the network hardware. Now what kind of company is going to turn on something like that when it allows people inside the network to control their hardware. None?

        When I heard what UPnP did, I was astonished and horrified, but I has a skim-read of the spec, and the standard does appear to support some form of authentication. In other words there is a mode of operation where authenticatedauthorised people inside the network can control the network hardware. T
  • by Anonymous Coward on Wednesday December 28 2005, @01:46PM (#14353151)
    Please add "Book Review: " to the beginning of the title. This is the second time I've noticed this.
  • by voice_of_all_reason (926702) on Wednesday December 28 2005, @01:46PM (#14353154)
    Get ready for it...

    Pay attention!

    Even if you're a Fortune 500 company with a 70-story building, you'd be surprised what a walkaround by the CTO can accomplish. Stick your head in a few cubes, say "what the shit is going on here?" and let the rumour mill work for you.

    It will take less time/money then hiring a "solutions" firm to police your internets. And it's the same way midlevel managers make sure their employees haven't been screwing around since like, forever.
    • Bah, screw that. Just block the ports on the firewall. If certain users need those services, then do a NAT directly to their workstation, and put that workstation on a subnet that can be isolated from the rest of your systems. Firewall based security isn't a total solution, but if you have a tight firewall then your security problems are so much more managable.

      I had a client who objected to this one the grounds that their employees used it "only" to talk to each other, so it was more "efficient" to keep the service. So I set them up a jabber server in the building, and blocked all outgoing traffic. The boss was fine with it, and while the employees were pissed as hell, they couldn't say anything about it because they'd all sworn that they weren't using it to chat with people outside the building.
    • by tpgp (48001) on Wednesday December 28 2005, @01:58PM (#14353233) Homepage
      you'd be surprised what a walkaround by the CTO can accomplish.

      You're right that this will stop a lot of problems - maybe even up to a third (and I generally agree that this is something a CTO should consider doing)

      However, it does nothing for:

      1) Malicious users (OK they're pretty hard to stop no matter what)
      and
      2) Stupid users who are using IM for legitimate company purposes, and get a message from their workamte / business partner saying "lol no this is not a virus." [slashdot.org]

      I certainly think companies should think about these applications in their security planning.
      • Blocking all but trusted attachments in email (.doc, .txt. and .exl okay, but not .doc.exe.vbs) should be alot easier than 1) making and 2) keeping current a blacklist for ports and numbers/users who are exempt. People who really need to transfer some wierd file type will find a way (put it in a zip, sneakernet), but it will cut down on viruses are malware, which -- because they can really muck things up -- require a more technological solution than harsh language.

        But most of these policies are to block
      • However, it does nothing for:

        1) Malicious users (OK they're pretty hard to stop no matter what)


        Um... maybe companies shouldn't hire malicious employees.
        • by steve_bryan (2671) on Wednesday December 28 2005, @03:33PM (#14353785)
          Um... maybe companies shouldn't hire malicious employees.

          Have you ever read any of the memoirs of Richard Feynman? I'm not going to make the ridiculous claim that every malicious employee is the equivalent of Nobel prize physicist Feynman, but any objective review of what he claims to have done makes it clear he would be classified as malicious. He found the security at Los Alamos labs during WWII to be onerous and pointless in the manner it was handled. That inspired him to various exploits that caused headaches for them. On the other hand he was one of the best physicists our country has ever produced. His contributions during the Manhattan Project might have been crucial. The idea here is that making the security department happy might not be the most important criterion when choosing employees.
        • In my experience, malicious users aren't hired. They are created by the company that employs them.
    • You need to read some books on management. Doing this will never, ever work. You'll just get people to periodically give a look to see if the CTO is walking around, which is even more of a distraction from work.
      • Not to substitute axioms for well-thought out ideas, but you can't catch all the fish in a pond anyway. Just because a solution isn't perfect doesn't mean it's not more valid than the current model.

        And if you have people slacking off in order to not get caught slacking off, then that's a whole other type of problem. Might be better to flush out the deadweights by doing it, actually.
      • That's a little harsh, eh? I mean, yeah, you have to know what the policy is before you push off into the deep waters of the internet, but zero tolerance always equals zero sense.

        I use firefox at home with adblocker. Lots of sites surprise me at work when I see what the ads are actually hawking. If I find one of them has teh boobies, then I can't go there anymore. No harm done.
  • PEBKAC (Score:3, Insightful)

    by SatanicPuppy (611928) <[Satanicpuppy] [at] [gmail.com]> on Wednesday December 28 2005, @01:47PM (#14353161) Journal
    I don't see how it is possible to secure an open protocol that allows file transfers. There is always going to be some idiot who'll click on the bad link, and download the trojan that can compromise the security of the entire network.

    Even if you put in multiple cutouts when dealing with untrusted users, inevitably you'll have a trusted user who will unthinkingly violate protocol and open the whole setup to exploitation.
    • Re:PEBKAC (Score:4, Funny)

      by killmenow (184444) on Wednesday December 28 2005, @02:20PM (#14353347)
      ...inevitably you'll have a trusted user...
      Sorry, you lost me right there.
    • Admin's problem (Score:4, Informative)

      by CarpetShark (865376) on Wednesday December 28 2005, @02:31PM (#14353405)

      Certainly, social engineering attacks come down to user education.

      BUT, there is NO excuse for not having the technical side locked down. It's all too common for people to claim that you can't protect against someone clicking on a link. The fact is, you CAN. Quite simply, install a secure browser (dump IE, in other words), put it through a filtering proxy like dansguardian, and then close http ports on the firewall, except for the proxy server itself. Disable webmail at the web proxy, and disable downloads anyway at the same proxy. If you need windows update or something like that to work, you can explicitly allow certain sites. But DON'T allow any more than strictly necessary. Don't allow SSL, except to trusted sites where no uploads or downloads or conversations take place.

      Likewise, install a secure email client, and have mail filtered through a company mail server, disable HTML mail and encrypted mail.

      These are basic security precautions. But already, you've secured your organisation far beyond most of the windows shops out there that get virus and spyware issues every day.

      It doesn't take a genius, it just takes you to choose what technology you allow on your systems, and to use it wisely.

      • Sure, if you can get away with all those things.

        Internet Explorer is still necessary for viewing some websites. I can't put in my damn expense reports without IE because the wankers who wrote the site wrote it using Microsofts Java, which only runs with microsofts crappy browser. All the management here uses Outlook, and corporate is migrating everyone to Exchange. They'd go nuts if we tried to take away their shiny HTML mail.

        We get tons of ads (ads that we get paid to publish) in email, generally pictures,
        • Even in IE, you can set which sites are allowed to do things, and which aren't. With both IE and Outlook, you can set proxies and filter mail so that you only allow stuff from trusted senders etc.

          But yes, if you have people above you who control IT policy, there's not much you can do, except make sure they take responsibility for bad decisions rather than you, and that you keep looking out for a better job. Admittedly, that can be hard to find too :(

    • Therefore you buy yourself a piece of software that can virusscan these files instead of blocking them ! Oh protocol xyz can be used to transfer files (name 1 protocol that cannot be used for this purpose ? even ping can be used to transfer files).

      "There will always be one idiot who" -> perhaps, but why punish 1000 non-idiots instead of firing the idiot ?

      If IT security becomes synonim with bullying (which it is in many companies), I can assure you nobody, absolutely nobody will care about security, and t
      • If you catch it in time, and if the thing you downloaded isn't capable of logging and trasmitting a password. But what if you don't, or it is?

        Right now worms and viruses are easy to spot, because the first thing they do is spam themselves out all over the place. Gives you tons of warning. But what happens when you get one that spreads slowly, under the radar? Then you've got a long term vunerability on the network.
  • I work as a security consultant for Hospitals and Banks. In some of the audits I have done, I have found that are controls, or even considerations for IM. Even P2P, I was at one bank that the VP of the place ordered the third party vendor to open ports on his firewall for P2P stuff. I am no legal expert, but I told the 3rd part to get it in writing from the VP (if he still wants it open after our scathing report) that the VP orders the 3rd party to open those ports. That way, the Bank and the VP are the l
  • by digitaldc (879047) * on Wednesday December 28 2005, @01:49PM (#14353177)
    The formatting uses a font size that is somewhat larger than other book. This seemingly serves to achieve a high page count.

    Is this a security book or a term paper?
  • by Ed Avis (5917) <ed@membled.com> on Wednesday December 28 2005, @01:51PM (#14353194) Homepage
    File transfer can still be carried out by most instant messaging clients, and that can pose serious security risks.
    I'm not convinced of this. It's not as if the instant messaging client magically runs with higher privilege and gives someone access to files they couldn't otherwise view. If they transfer a file to a friend, it must be a file they already had permission to read. If they receive a file by instant messenger, the risk is no greater than if they'd simply downloaded it in their web browser or loaded it from a CD.

    I'm deliberately taking a one-sided position here, but it seems there is a lot more heat than light generated over file-sharing 'dangers'. I am reminded of Catbert's banning of camera phones as a security risk - notwithstanding the fact that the only documents people could take photographs of would be those they're allowed to read and photocopy anyway - and without even banning ordinary cameras.
  • by Bullfish (858648) on Wednesday December 28 2005, @01:52PM (#14353197)
    This may help some companies get an idea of what all activities going on in their network, but I doubt anyone will ever stop the activity going on as described. For companies, the biggest deterrent will remain getting fired if someone is using work computers to do P2P or IM. If the company policy is clear, and people are aware of it, the company really only has that (and a series of graduated warnings) to use as a club. Blocking ports, trying to shape protocols, trying lockouts etc are, IMHO, a waste of time. A workaround will always come. Better to have a clear policy and enforce it than buying fancy-ass software or spending 50 bucks for a book on what any good IT manager knows already.

    Out in the world of ISP's (which is different than in companies), the same situation exists. Try to block P2P, or bittorrent, and someone will find a way around the security. They could kick people off their service driving them to another ISP, but that's about it. This book doesn't really sound like it applies to that situation really.
    • I could not have stated it better than you. You are at work, you are being paid to work, not P2P or IM. Do it and you are fired. Where I was, we have this policy as well as not allowing users to install their own software. No P2P software or IM software is originally on the machine, if it shows up, the machine gets formatted as so does your employment.
  • ....is Eric Rescorla's SSL and TLS: Building Secure Systems [amazon.com]. It's got excellent descriptions of how SSL works, including a chapter on various attacks (million message, small-subgroup, etc). He's got some nice stuff in chapter six about SSL server performance, too - talks about hardware acceleration and whatnot.

    Oh, and, plug [pmdapplied.com]!
  • by everphilski (877346) on Wednesday December 28 2005, @02:01PM (#14353248) Journal
    with the majority of such titles being in the $25.909 - $39.99 range

    When $25.90 just isn't enough, but $25.91 is just too much...

    -everphilski-
  • by Twid (67847) on Wednesday December 28 2005, @02:07PM (#14353282) Homepage
    I guess the flagpole metaphor would make sense if a flagpole was a security device.

    I think what he is trying to say is that there is no use putting a gate on your driveway unless you put walls around it as well. Otherwise people will simply drive around the gate.

    Certainly works better than the flagpole story anyway, unless there's a secret security use for flagpoles than I am missing. :)
  • Someone else posted the obvious solution of blocking all the ports at the firewall. That's simple enough, but stupid people can still download software via the web and mess things up.

    The simplest solution is to lock down the user's rights. Just prevent them from installing any software and don't put P2P or IM clients on their systems. Problem solved. If you really need them to be able to use IM, run it via MSN IM through your Exchange server (I'm sure there's OSS alternatives to do the same thing). Tha
  • I already know how to secure IM and P2P apps, so this book, imho, sounds like a complete WASTE [sourceforge.net] of time.

    nudge nudge wink wink...say no more...
  • Yeah, businesses really need this so their employees can securly trade dvd torrents and chat with their spouses.

    • Mod parent up- this is the real thing. Synchronous communications have a tendency to be anti-productive. Asynchronous communications have a tendency to be productive up to a point, and easily ignorable if an emergency happens that needs to be responded to. Based on this idea, I'd also suggest removing phones from cubicles, as well as not allowing IM and P2P applications be installed in the first place.
  • Why would you assume these IM/P2P applications are even installed in the first place?

    In most corporate environments, software policies are already in place to restrict users from installaing any software on their own. In addition, generally any requests for installation of IM/P2P apps are quickly denied citing company policy (the reasons for which should be painfully obvious).

    There's really no need for IM at work, but if you really really want it, use a corporate IM solution (such as Exchange IM or Apple iC
    • Re:False assumption (Score:4, Informative)

      by slim (1652) <john@haCOUGARrtnup.net minus cat> on Wednesday December 28 2005, @03:21PM (#14353698) Homepage
      There's really no need for IM at work,

      I work in a corporate environment with geographically diverse colleagues, and IM is an extremely useful medium for doing Real Work. You might like to argue that we could just as easily use the phone, but IM has advantages over the phone for certain applications. Especially, it's nice to be able to supplement phone conversations with IM -- we'll cut and paste email addresses, code fragments, log fragments, even screenshots rather than try to read them out or describe them.

      On telephone conference calls, IM is a useful out of band medium for comparing notes with colleagues; "Should I mention x?", "Don't forget y". ... but if you really really want it, use a corporate IM solution (such as Exchange IM or Apple iChat) to keep things local. Problem solved.

      I agree with this. OTOH, it's in my employer's interest to allow me access to MSN messenger. Some of my technical peers work for different companies. If I have external IM, I can go to them for technical assistance (and they can come to me: it's a two way street).
  • Yet another analogy (Score:4, Interesting)

    by dtfinch (661405) * on Wednesday December 28 2005, @02:42PM (#14353461) Journal
    Putting "protected by [insert alarm company name here]" stickers on the windows of my house will discourage most of the amateurs from breaking in, even if I don't really have an alarm. Even the pros may skip to the next house without looking, unless they know I have something they want. Not that I condone improper use of cryptography or anything, but you can use analogies to support any position.
  • I'm surprised at this point that nobody's mentioned just how bizzare the topic is in some sense.

    I mean, Securing IM is a legitimate and important thing for corporate IT departments and people with real responsibilities to concern themselves with.

    On the other hand, "Securing P2P" is basically just another step forward in the arms race between those who would choose to flaunt copyright laws and those trying, however vainly, to stop them. Even if you would try to make the rather weak case that P2P has leg

  • by Rurik (113882) on Wednesday December 28 2005, @04:28PM (#14354106)
    Thanks for the brief, though good, review of the book. I'm B.B., author of three chapters of the book: 9, 11, 12 (Gnutella, BitTorrent, FastTrack). If you look at the book's profile on other sites, you'll see there were a variety of co-authors on the book. As a long time member of Slashdot, and a long time advocate of both Open Source applications and Linux, this was a small way for me to at least give a little back. My chapters were written from a Linux admin POV, with details and steps on iptables (with installing strings), self-made Snort rules, and Ethereal screen shots (which were done in Windows, my Linux boxes are headless) I can only speak for my sections, but I hoped that if a regular Windows admin picked it up, and saw how easy it was to create firewall rules in Linux, it may help to win some hearts and heads.

    Overall, it was an honor and priviledge (cliched, I know) to help out with the book, with a great bunch of other guys. And thanks Slashdot ;) //Feeling obliged to use Karma Bonus
  • Ironically (Score:3, Informative)

    by Orion Blastar (457579) <orion_blastar@NOspam.yahoo.com> on Wednesday December 28 2005, @05:00PM (#14354259) Homepage Journal
    most companies that try to lock down their Internet programs often use Internet Explorer and Microsoft Outlook as the default web browser and email client. Yet these two programs have the most exploits of any Internet based programs out there. So even if you do lock down the ports of the firewall and stop users from installing programs, chances are the exploits will install the malware for you when they get the wrong email or click on the wrong link.

    99% of the malware infections that happened in the past four places that I worked in, were caused by management clicking on the wrong email or wrong link in Outlook or IE. They did lock down their Internet, turned off port forwarding, took away admin access, prevented the install of new programs (which screwed up Visual BASIC and MS-Access development, because they needed Admin access or else things don't work via certain controls), and other things.

    I think one of the funniest momments was getting the "Love Bug" email from the Network Administrator 12 times in a row that said "I LUV YOU!" over and over again. Guess who was using MS-Outlook and McAfee Anti-Virus and got infected due to some exploit? Needless to say I was smart enough not to open up those emails, unlike my co-workers who did, and sent me their own "I LUV YOU!" emails. :)
    • I agree. Too often admins see the problem of "insecure or unwanted traffic on port XX" and solve it by blocking port XX. My question is why wasn't that port already blocked? As a system administrator I block All ports except the ones we need. Even then those ports are monitored for the correct kind of data.

      No this won't stop all the baddies, but why would you leave ports open at all?