Fatal System Error

brothke writes "As computing and technology has evolved, so too have the security threats correspondingly evolved. The classic Yankee Doodle virus of 1989 did minimal damage, all while playing a patriotic, albeit monotone song. In 2010, aggressive malware now executes in stealth mode, running in the background with an oblivious end-user, and antivirus software that can’t detect it." Read on for the rest of Ben's review. Fatal System Error: The Hunt for the New Crime Lords Who are Bringing Down the Internet author Joseph Menn pages 304 publisher PublicAffairs rating 8/10 reviewer Ben Rothke ISBN 978-1586487485 summary Non-fiction cyber-thriller with super analytical advice Cybercrimes have evolved using increasingly sophisticated techniques, and the resulting financial losses are staggering. Many criminal cyber gangs are well organized and resourceful and their ability to recover after new defenses have been deployed make it a challenge for those on the right side of the law.

Fatal System Error: The Hunt for the New Crime Lords Who are Bringing Down the Internet is an excellent book billed as a non-fiction cyber-thriller, and describes the cyber gangs who operate on the Internet. Author Joseph Menn, a cyber security reporter for the Financial Times, takes the reader into the inner operations of today's cyber-criminal, who use the Internet as their personal mint.

While Willie Sutton never really said that the reason he robbed banks is because that's where the money is; the truth is that today's cyber criminal does know where the money is, and its address is the Internet. They use the net as a means to steal and extort money from businesses and individuals.

The book's protagonist is Barrett Lyon, a highly skilled technical engineer and entrepreneur, who founded companies such as Prolexic, BitGravity and 3Crowd. It was at Prolexic where Lyon developed the software used to fend off the DoS attacks that were bringing some of his client's networks to a standstill.

Lyon, along with the other major character in the book, Andy Crocker, a British policeman, were the 1-2 punch that resulted in the prosecution of a Russian cyber criminal. The fact that the prosecution took place via the Russian judicial system was a surprise to everyone. What was unusual about the prosecution is that criminals in Russia and Eastern Europe often operate with the assistance of corrupt political and police forces. Even though the evidence against the defendant was significant, the ability to secure a guilty verdict was far from a sure thing.

Much of the book deals with Lyon and his working relationship with BetCRIS, a company offering online gambling services, including sports betting, online casino games, online bingo and mobile gambling.

BetCRIS is an off-shore company, operating in the safe havens of the Republic of Costa Rica. In 2003, at the height of the DoS attacks, the BetCRIS website was down for nearly a month. With tens of millions of dollars of gambling revenue at stake, BetCRIS management were desperate for a solution, and they reached out to Lyon.

While Lyon created a first-generation solution to stop the early DoS attacks, the book details how the attackers were able to get around those countermeasures, and how it turned into a cat and mouse game of futility, where Lyon would create a fix, only to be beguiled by a new attack.

In the book, Menn writes about many of the major players in the Internet criminal world. He spends a good amount of time writing about the infamous Russian Business Network (RBN). He notes that little true business was carried out via the RBN; rather it was a front for Internet-based criminal activities in Russia.

Menn does get into some technical details, but not so much so to confuse a non-technical reader. He covers topics such as botnets, DoS and DDoS attacks, cyberwarfare, cyber espionage, and the difficulty in prosecuting the perpetrators.

Menn notes that there are many reasons why Russia and in Eastern Europe are ground zero for cybercriminals. The educational institutions there provide a good source of technical training; combined that with the fact that legitimate job opportunities are often quite limited. Add to the fact that political and law enforcement officials often ignore the cyber attacks again the rich capitalists of the US, the difficulty and challenges with jurisdiction, and you have a perfect storm for the creation of a sophisticated cyber criminal element. Finally, there is a long and established culture of corruption in Russia and in Eastern Europe that adds to the problem.

There are two directions that Fatal System Error takes. The main part of the book is Menn's narrative, which takes up 11 of the book's 12 chapters. These 11 chapters take the reader on an enthralling ride into the inner workings of the cyber-criminal world. Fatal System Error is an enjoyable read on par books such as The Cuckoo's Egg: Tracking a Spy Through the Maze of Computer Espionage and Takedown: The Pursuit and Capture of Kevin Mitnick.

Where the book truly stands out is in the final chapter Fixing What's Fixable, and is worth purchasing for that chapter alone. Menn displays his incredibly deep understanding of the underlying issues around computer security and why we are vulnerable. He suggests numerous pragmatic solutions to the crisis, and how to better secure the Internet and networks.

Some of the ideas include significantly greater budgets for information security, more liability against software developers who write insecure code, greater information sharing between the cybercrime agencies in the US and their counterparts in Russia, and more. His on-target analysis of what the US Government can and should do to increase the security of the Internet infrastructure is quite impressive.

Reading the narrative part of the book, many readers will likely be scared to death to connect their computers to the Internet, and to a limited degree, rightfully so. Even with Menn's balanced and compelling account of what transpired, the threat of identity theft and ease of how financial accounts are breached may be too much for some readers many to bear.

If corporate America and the US Government would take Menn's suggestions to heart on how to create a secure Internet infrastructure, many of those security concerns he wrote about could be obviated, and the cyber criminals of Eastern Europe would have to look for different work.

Additional pragmatic ideas that Menn suggests are to legalize and regulate online gambling, more funding to teach safer computing in schools, and for a complete re-engineering of the Internet, in order to build in the necessary security functionality which should have been in there in the first place. As part of the process to re-engineer the Internet, Menn suggests designs that create accountability into the Internet fabric.

Finally, Menn notes that many end-users are not blameless. By not educating themselves on how to securely use the Internet, they are setting themselves up to becoming victims. He writes that anyone that connects a computer to the Internet needs to have significant security vigilance to ensure that they don't make themselves a victim. It is 2010 and far too many people are still oblivious to the security threats. Many still naively believe that someone from Nigeria really does want to make them richer with tens of millions of dollars worth of gold from their deceased uncle.

Menn shows how the underlying infrastructure of the Internet is significantly more vulnerable than most people realize. Finally, what exacerbates the problem is that those doing the attacks are working much quicker than those who are trying to secure it.

One of Menn's criticisms is that the US Government spends a fraction of what it should on securing its critical technology infrastructure. Fatal System Error: The Hunt for the New Crime Lords Who are Bringing Down the Internet is the wake-up call that those in Washington, and those charged with IT need to wake up to. Unfortunately, it is likely those that truly need to read this book, will press the information security snooze button yet again.

Ben Rothke is the author of Computer Security: 20 Things Every Employee Should Know.

You can purchase Fatal System Error: The Hunt for the New Crime Lords Who are Bringing Down the Internet from amazon.com. Slashdot welcomes readers' book reviews -- to see your own review here, read the book review guidelines, then visit the submission page.

Comments

Re:No Impartial Book Reviews

[ProdigyPuNk]

While I would usually agree, the amazon reviews suggest that this one really is a pretty good book

http://www.amazon.com/Fatal-System-Error-Bringing-Internet/product-reviews/1586487485/ref=cm_cr_dp_all_helpful?ie=UTF8&coliid=&showViewpoints=1&colid=&sortBy=bySubmissionDateDescending

Uh, no

[causality]

Additional pragmatic ideas that Menn suggests are to legalize and regulate online gambling, more funding to teach safer computing in schools, and for a complete re-engineering of the Internet, in order to build in the necessary security functionality which should have been in there in the first place. As part of the process to re-engineer the Internet, Menn suggests designs that create accountability into the Internet fabric.

Great, so they want to redesign the Internet because people don't want to learn how to identify a phishing site and can't understand that giving your account numbers to unverifiable strangers is a bad idea? No thanks. A fool and his money are soon parted and there's not much you're going to change about that. Also, I'm sure that "accountability" is a euphamism for "tracked everywhere you go even more than you are now". Seems to me they are trying to increase protection against petty criminals while drastically reducing protection against overzealous governments that want to censor.

Re:Uh, no

[caffeinemessiah]

Great, so they want to redesign the Internet because people don't want to learn how to identify a phishing site and can't understand that giving your account numbers to unverifiable strangers is a bad idea?

Oh please, I think Sony put an end to the delusion that only grandmas and morons are susceptible to phishing or malware. Allow me to give you an example which most people here won't be able to do detect instantaneously: zero-day exploit in Flash + rootkit + trojan. I run a tight ship like the next nerd, but my AV software still flags trojans that somehow make it onto my system from time to time, and those are only the ones that it CAN detect.

And yes, there are zealots who will undoubtedly say things like "Flash is for suckers" or "what do you expect with Windows?", but these people should consider the fact that (a) not everyone lives in caves, and (b) some people just have more important things to worry about, like losing their homes.

Re:Uh, no

[rickb928]

"Great, so they want to redesign the Internet because people don't want to learn how to identify a phishing site and can't understand that giving your account numbers to unverifiable strangers is a bad idea?"

Uh, no. With MITM attacks, spoofing raised to a fine art, SSL hijacks of any number of diffeent methods, fake/spoofed/stolen certificates, it can be very, very hard to avoid making a mistake and trusting something you should not.

"No thanks. A fool and his money are soon parted and there's not much you're going to change about that."

Doh.

"Also, I'm sure that "accountability" is a euphamism for "tracked everywhere you go even more than you are now". Seems to me they are trying to increase protection against petty criminals while drastically reducing protection against overzealous governments that want to censor."

Yup. It's a fine line between security and oppression.

Re:Uh, no

[causality]

Great, so they want to redesign the Internet because people don't want to learn how to identify a phishing site and can't understand that giving your account numbers to unverifiable strangers is a bad idea?

Oh please, I think Sony put an end to the delusion that only grandmas and morons are susceptible to phishing or malware. Allow me to give you an example which most people here won't be able to do detect instantaneously: zero-day exploit in Flash + rootkit + trojan. I run a tight ship like the next nerd, but my AV software still flags trojans that somehow make it onto my system from time to time, and those are only the ones that it CAN detect.

And yes, there are zealots who will undoubtedly say things like "Flash is for suckers" or "what do you expect with Windows?", but these people should consider the fact that (a) not everyone lives in caves, and (b) some people just have more important things to worry about, like losing their homes.

Flash is known insecure software with a terrible track record, and I treat it as such. I obviously can't make others do the same but they're crazy not to. It undoubtedly helps that I am not using Windows (just why that helps is a separate debate). That to me is basic common sense combined with a few minutes of Googling. If that's the standard now for "living in a cave" then the standards these days are quite low. For your item "b" there, it's a lot easier to keep your home when some criminal hasn't drained your bank accounts for you.

It's not about Flash, Windows, living in caves, or having other concerns in life. No, those are all distractions from the actual issue, and you can tell because they're always said in the same irritated emotional tone. It's about two different mentalities. They come up in lots of otherwise unrelated issues including those that are much more political in nature. One mentality wants to look after its own interests and equip itself in order to protect itself. The other believes that is too much of a bother, not their problem, or otherwise is someone else's job. I do not exaggerate in the least when I say that big government of the "we know what's good for you" variety derives most of its existence from the latter because these people want someone to take care of them, almost like children.

So I secure my systems after teaching myself how to do so, and I study good practices. Another person thinks this is too much of a bother and goes with whatever vendor defaults his system came with because to him, security is that vendor's problem only. Guess who gets compromised? Which do you suppose is an easier target? It's not about time or any of those other excuses because you always have time for something you consider important. "I don't have time" is a cute way of saying "this is not a priority". It's about personal responsibility and whether you realize that no one wants to protect your interests quite as much as you do, that all the tools and information you need are out there. Do I have time to be personally responsible and take only the amount of risk I want to take instead of being helplessly dependent on someone else to protect me? Yes, I do have time for that, no caves required.

Re:Uh, no

[causality]

Uh, no. With MITM attacks, spoofing raised to a fine art, SSL hijacks of any number of diffeent methods, fake/spoofed/stolen certificates, it can be very, very hard to avoid making a mistake and trusting something you should not.

I agree that there are sophisticated methods by which a determined adversary concentrating his efforts against a particular target might effect a compromise. However, if all compromises were of this type only, then ID theft would be a nearly unknown crime and botnets unheard-of. That's because an attack this effective and sophisticated does not easily lend itself to automation. Criminals can't compromise tens of thousands of machines or build large lists of account numbers that way. If the level of sophistication you mention were the bare minimum requirement to break the security of the average user, we'd have a global Heaven on Earth in terms of network security.

The vast, vast majority of phishing attacks are quite crude by comparison. They are crude because crude works. People fall for it, all the time. These simpler, less sophisticated attacks are easy to automate and send to thousands of users. When a criminal can send a simpler attack to many thousands, only a small percentage need to get suckered for him to profit immensely. This is where most of the problems are coming from, not dedicated personal efforts against specific targets that require a lot of manpower and expertise to execute. I think the latter is within the realm of statistical noise by comparison.

Yup. It's a fine line between security and oppression.

Security of the "be responsible for your own host and your own network because it's in your interests to prevent their compromise" doesn't lend itself to oppression. Security of the "we will be the central authority who will do everything for you" variety certainly does, not to mention it probably won't work. I think when it comes to security, it's perfectly reasonable to say "if you don't care, neither should we." It really doesn't take much to be a much harder target than the lowest-hanging fruit.

Re:Uh, no

[causality]

I believe the Ubuntu screensaver issue was from the Gnome-Look.org site, not the official repositories. My apologies if you're referring to different virus I have no knowledge of. That said, you are correct, unix and linux are not remotely immune.

The difference is that *nix systems in their various incarnations have had decades of exposure to all sorts of attacks and have evolved accordingly. I would not call them immune, I would call them resistant. There are many good tools available to secure them and, unlike Windows, these tend to be considered standard system utilities not third-party add-ons.

I believe the whole "anti-virus, anti-malware" mentality of removing an infection after a compromise has taken place is fundamentally broken as a security measure. That's because it is not security at all; it is damage control. After your security has failed it might be useful for containment but that's about it. The correct way to respond to a system compromise is to format the drives and reinstall the OS from known good media. Real security systems are designed to prevent compromises, not to remove malware after a compromise has happened and malware has been installed. This is what you find on *nix. It's not just systems and tools, it's a mentality that goes with them.

This is why there tend not to be successful viruses (I use the term loosely to also include worms and such) propagating in the wild on *nix systems. There do exist viruses for *nix systems; they're called proofs-of-concept. Like all self-replicating forms of malware, they have something in common: they must compromise the system (either a user account or root) before they can do anything else. That is what *nix systems are good at preventing. It also helps tremendously that *nix systems tend not to be the "write once, compromise millions of machines" monoculture that you find on Windows.

The last thing I'll say is that average *nix users tend to be more competent and more knowledgable than average Windows users. They're more likely to know a risk when they take one. They're more likely to understand why Flash and other software with a terrible security track record is not trustworthy and should be treated as such. They tend to have habits that reduce their exposure. Overall, they are harder targets and don't represent the low-hanging fruit. None of this amounts to "perfect immunity" of course, but represents a hell of an improvement over the average.

Re:Uh, no

[element-o.p.]

Pretending that unix/linux solutions are inherently safe is about the same as sticking your head in the sand. The only reason they're safer is because people aren't that interested in exploiting the relatively few people who use them.

Inherently safe? Yeah, you're probably right. Even the best, most secure OS in the world can't protect a truly motivated idiot from himself. Inherently safer , however, is what I would claim for Linux, based upon my own anecdotal experience. It's harder to hose an entire Linux box than an entire Windows box and easier to clean up after the fact (having had to clean up both OS'...YMMV). I knew a Linux admin (and I use the term very loosely) who constantly had his boxes hacked on a regular basis. As a result of his experiences, I took a really long, hard look at my choice of OS to see if it really was that much better. In the end, I realized the guy was just an idiot. He didn't take basic precautions to secure his machines, and he was regularly exploited because of it. The one other Linux box I have ever seen compromised was a public-facing FTP and web server where a user with a weak password had their account compromised and PHPShell was uploaded into their public_html directory, and was used to install a spam relayer. That was the extent of the exploit -- they never gained root privileges, they never got outside of the user's home directory. It took us two hours to detect (because it was hacked one hour before we got to work), and maybe another hour to clean up. Cleaning up Windows infections is a whole other story, and that's why most home users just buy a new PC when they get a virus. They run A/V and if the problem is still there, they throw it out and buy a new PC. That's wasteful and expensive.

Like it or not, Windows is the premier operating system in the world, for personal computers.

For now, yes. But that's changing. Like it or not, Linux is becoming more mainstream. My entire datacenter, except for three servers, is Linux (well, and one FreeBSD-based appliance). *ALL* of the desktops our field personnel use are Linux. You can now buy Linux installed from mainstream OEMs.

The average user is never going to be a linux nerd...

If by "Linux nerd" you mean someone like me, who likes to build and maintain Linux machines, yeah, you're probably right. But the average Windows user is not a "Windows nerd", either. They aren't downloading and installing beta versions of the next Windows OS, they aren't tweaking their system on a daily basis, and they don't rebuild the OS when their computer stops working. The average computer user just wants software that works, regardless of operating system. They don't really know the difference -- or care -- between Windows and Linux, because they only want to surf the Internet, send e-mail and type up the occasional document or spreadsheet.

...the OS is unstable...

Pot...Kettle...Black? I've got Linux servers and routers that have uptimes of over five years (well, I did...a few recent power outages that outlasted the aging UPS's changed that). I have *never* seen a Windows box with an uptime like that. Again, YMMV, and the plural of anecdote != data, etc., etc.

...and requires too much technical expertise with too little software support. You can deny that, but we all know it's true.

Yeah, I'll deny that. Ubuntu has become at least as easy to install as any version of Windows I've ever used. Once it's up and running, it's not any more difficult to use than Windows, and speaking as a sys admin who has to maintain about 70 Linux desktops that sit 500 miles away from my office, I can tell you it's far, far easier to maintain a Linux machine from the CLI across a satellite hop than a Windows machine using rdesktop or VNC. Even for a home u

Re:Uh, no

[causality]

I managed a very small ISP for a while on the 90s, and have my own mail and web servers to this day.

The definition of 'lowest-hanging fruit' for all the attackers out there is much broader than you implied. If you have a host accessible via the Internet, you ARE a target. You are being attacked now, this very minute. That you deflect those attacks ahead of the host at firewall, router, or application level doesn't change that. It just makes your logs bigger or smaller.

Your operating system choice makes no difference. They attack everything. You just use different tools and methods depending on what's available and what works.

I know what you mean. I run a very small-scale personal-use SFTP server (no shell access for any account) so I can access some of my files remotely. I use SSHGuard to hinder brute-force attacks and LogSentry to keep abreast of the activity. I constantly receive attacks at all hours of the day. They're quite dumb and have little or no sophistication; most are just trying to guess default passwords for system accounts and such.

I have told many people the same thing you just said. I have explained that if you run any sort of Internet-facing network service, you will get attacked and probably with high frequency. There is no such thing as "so obscure and small-scale that you're under the radar". Expect it and plan for it. The people who are surprised when this happens are the easy targets.

I disagree that my choice of OS makes no difference. I submit that my Gentoo Hardened system with very strict security policies is more difficult to compromise than a Windows installation on the same hardware offering the same SFTP service. When you build everything from source, you can implement protections against buffer overflows and other vulnerabilities that aren't available on a closed-source OS. With a *nix system, the tools I am using are not some black box. I can take them apart, examine them, and really understand how they work before integrating them into my system. The system itself is transparent. If something goes wrong, I can always find out why and can almost always do something about it. If something breaks, it broke for a good reason, it'll stay broken until I fix it, and when I fix it it'll stay fixed. My experience with Windows has been nothing like this.

I am not saying that one cannot run a very secure Windows system. I am saying it's easier to achieve the same level of security with a *nix system. More than that, it's easier to actually understand what you are guarding against and why your measures are effective. I think the importance of that last point is underappreciated. It cannot be properly appreciated in the realm of "run this anti-malware product and hope it takes care of things for you" and the mentality that goes along with it.

Yankee Doodle is not monotonic.

[Anonymous Coward]

There's 3 different tones in the first 4 notes alone, goodness!!!

Pedantic, but...

[Anonymous Coward]

monophonic != monotone

Re:Pedantic, but...

[CannonballHead]

That's not pedantic, that's basic terminology. MonoTONE would be one TONE. Monophonic would be one "sound" [at a time]. The "monotonic Yankee Doodle" does not even make sense...

Reviewer confused about Slashdot

[Areyoukiddingme]

Somehow it appears the book reviewer confused Slashdot for the Ladies Home Journal. Was it really necessary to use the "cyber" prefix 47 times? Really? Because we're so impressed when it's a cybergang, instead of just a gang.

One hopes the book isn't that bad...

Frank cyber-walked his cyber-beat.

[khasim]

He knew there was a cyber-gang out there waiting to commit their next cyber-crime. Frank knew he had to catch them with the cyber-goods. Frank's 45 wouldn't be much help on this cyber-collar. Frank needed something better. Frank needed a cyber-45. Frank knew only one person who could supply him with that, Cyber-Jimmy. The best cyber-fence in the cyber-world. Frank pulled up to the next cyber-phone to give Cyber-Jimmy a cyber-call.

The cyber-phone cyber rang.

Cyber-Smurf here, came the reply.

Fear and loathing in PC Town

[gx5000]

All i see is another book that uses paranoia and fear to sell....
Hidden code...oooo....Stealth Mode executing..aahhhhh...Root kits ! *GAG*

I know we're talking about the common user here....
But drive a car with no regards and you get the same thing...an accident.
Get a mechanic, a good one that can show you the pratfalls and some fixes.

But if you drive like a fool and visit "those" sites you get what you get.
Get Acronis a re-image your ass every week....you'll be fine.

Missing something?

[American AC in Paris]

"As computing and technology has evolved, so too have the security threats correspondingly evolved. The classic Yankee Doodle virus of 1989 did minimal damage, all while playing a patriotic, albeit monotone song. In 2010, aggressive malware now executes in stealth mode, running in the background with an oblivious end-user, and antivirus software that can’t detect it."

Yeah, the 1989 Yankee Doodle virus was pretty harmless.

You need to go all the way back to 1988 to find a worm which effectively shut down the Internet.

How one can overlook the Morris Worm in this context is completely beyond me.

He calls for the end of OSS

[wiredog]

more liability against software developers who write insecure code

So now we have to buy expensive insurance before we write OSS code? What about the liability of students?

Flying Under the Radar

[nuckfuts]

malware now executes in stealth mode, running in the background with an oblivious end-user

I've long need puzzled by malware that doesn't do this. Many trojans I've cleaned from people's computers download other pieces of malware. I once gave a demonstration of "drive-by" infection where merely viewing a malicious web page on an unpatched system resulted in nearly 20 new processes being spawned in the background. Impressive, in a way, but exceedingly obvious. Even clueless users can't help but notice that something is wrong, and IT gets called in to clean it.