8 ms·
I'm kind of torn on this. On the one hand, yeah, it is a trivial flaw. On the other hand, so is waving a gun at a teller. That attack has been around for dec
by patio11 15y ago
I'm kind of torn on this. On the one hand, yeah, it is a trivial flaw.
On the other hand, so is waving a gun at a teller. That attack has been around for decades and still works a few dozen times a year, because the cost/benefit analysis says that after hardening the banks a little it is easier to just lose a few tens of thousands of dollars every once in a while than it is to give them the Secret Service's attention to physical security.
That is hardly the only systemic vulnerability in the banking system. For example, let's suppose I want to compromise your account number and credentials sufficient to take you for every penny you possess. You know what I need? A check of yours. Any will do. Everything I need to create a demand draft against your account is on every check you have ever written. Every employee of every business you have ever paid by check got the keys to your financial kingdom.
You may not be aware of it, but since those credentials are assumed compromised, the security is in a) catching me when I use the demand draft to suspiciously drain your account and b) failing that, making you whole out of the bank's pocket. The numbers have been crunched: it is vastly, vastly more efficient to treat fraud as a cost of doing business than it is to tighten the screws 100%
The attack surface on software the size and complexity of a bank's is like the Death Star, except any single rivet being out of place will eventually result in this headline.
(Step #1 in tightening the screws would be turn off public facing websites, because inexpert users plus compromised machines means that no banking website will ever be secure, even without coding errors. This will never happen, because the provable cost savings of moving customers to online banking roflstomp over the marginal fraud risk.)
- st3fan 15y agoThe attack surface on software the size and complexity of a bank's is like the Death Star, except any single rivet being out of place will eventually result in this headline. I agree that banking software is probably complicated. But really, this bug is a beginners mistake and it shoul dhave never happened. It is something that the original developers should have known about and also something that the company auditing this code should have seen. There is really no excuse for this.
- bigohms 15y agoThis is like the first thing any new hax0r tries. Imagine the joy when it actually worked...
- RMacy 15y agoseriously, it really is.
- int3rnaut 15y agoI was thinking the same thing--finally something my grandma and me could have hacked during some "grandma-grandson time".
- tptacek 15y agoI've been doing software security since 1994, but spent 10 of those years as a dev, and so didn't do my first web pentest until '05. On my first ever web gig, I scored a login prompt with 'OR''=' SQLI in the password field. It's like the White Whale for me now; I haven't seen it again, but I know it's out there. Arrrrrhh.
- bigohms 15y agoI can understand and even respect a good pw SQL inject, but they MUST HAVE sat there with pins and needs, giddy, saying to themselves: "It CANT be THIS easy!".
- damncabbage 15y agoI found exactly the same thing in the first week of my previous gig. Your White Whale unfortunately pops it head up everywhere.
- dclowd9901 15y agoThere's also something to be said for having more deterrence than the next guy. Which is better business? Cutting ridiculous corners to save on developing decent security, and eventually getting egg on your face, to the chagrin of your customers, or simply keeping idiotic measures like this out of your software so it's one of your opponents' faces that are hit with the egg?
- alex_c 15y agoNonetheless, banks still lock their branches at night. That's about as basic as security gets.
- ctdonath 15y agoThis isn't akin to waving a gun at a teller. This is akin to handing the teller a huge stack of withdraw forms with random account numbers on them, and the teller dutifully checking each one, ignoring the ones with invalid numbers (most of them), and handing over the account contents for all the valid ones. This doesn't work, because the teller acts as security: if something is amiss in the implied security checks, to wit there is no reason one person should be submitting requests regarding lots of account numbers esp. when most of them are invalid; she'll have alerted management & security within seconds of seeing the bizarre request. Your subsequent analogy/justification/complaint is only valid if ONE doctored URL were used. My bank, Chase, does in fact implement security against such a "lots of random account numbers" attack: not only must the account match, but the MAC/IP address, browser/cookie, and other under-the-hood identifiers must line up; any mismatch between account number and access tools initiates emailing or texting a verification code to a known address/phone, which then must be submitted to close the loop of verification and, only then, allow access. Not running some kind of "one account per access device" sanity check is insane. It's not about one rivet being out of place - such vulnerabilities are understandable. It's about having an uncovered vent lead straight to the reactor core - that's stupid.
- vnchr 15y agoLike the Death Star...
- jasongullickson 15y agoI think you're missing the point (or I'm inferring it incorrectly), that, put simply, there is a cost associated with implementing the security measure (both to build it and the inconvenience to customers) which in some cases outweighs the potential cost of the exploit.
- ctdonath 15y agoOf course cost/benefit ratios should be considered. Not much point to debating that when it's obvious they weren't. The lock on the front door was good, but the only thing that protected the bank until now was that nobody tried the windows & safes on the assumption that they would, in fact, be locked. That's not weighing costs, that's being criminally negligent.
- speleding 15y ago> The attack surface on software the size and complexity of a bank's is like the Death Star Really? I'm sure there is more to it then I imagine but I've used the websites of 3 different banks and their functionality is quite limited, it looks like something you could do in a Rails app in a few weeks. None of the real payment processing is done by the web app as far as I know, it is just a thin layer on top of a database with the transaction history and can store payment orders for batch processing later on. None of the real payment processing systems would face the internet I imagine, so I think the attack surface would be quite small.
- awj 15y agoRunning cost/benefit on this is great, except that the numbers still don't add up. Instead of a user id that is obviously some row in a database, use a UUID and do your db lookups against that. Worst case you have to put in some kind of local UUID -> user id lookup table because you can't manage to change the user record database directly. Now, barring a broken UUID generation scheme, it's virtually impossible to crawl for user accounts. Plus you have a scheme for providing the same guarantee on any other identifiers. Obviously that doesn't prevent the larger problem of "guessing your user id and munging a url lets me into your account without a login", but it at least makes the guessing part worlds harder for relatively little work.
- rmc 15y agoOn the one hand, yeah, it is a trivial flaw. On the other hand, so is waving a gun at a teller. Not really. To make it so that waving a gun at a teller doesn't work you need to do lots and lots of complicated hard things. However it's as hard to ensure that person with account A doesn't access account B.