This article is more than 1 year old

Security gumshoes locate source of mystery web compromise

How dunnit plot gets scripted

Updated The source of the mystery infection of more than 10,000 websites back in January has been uncovered.

Thousands of legitimate websites were compromised at the start of the year to serve up malware.

It seemed that the exploitation of SQL Injection vulnerabilities was involved in the automated attacks. The precise mechanism was unclear until earlier this week when security researchers discovered a malicious executable later linked to the attack on a hacker site.

The hacker utility used search engines to find insecure websites that it then tried to exploit using an SQL injection attack. The exploit included an SQL statement that tried to inject a script tag into every HTML page on the website.

The tool - which had an interface written in Chinese - was programmed by default to insert a tag to the same malicious JavaScript file that featured in the January attack, solid evidence that it was at least partially behind the assault.

The tool runs a script called pay.asp, hosted on a server in China. This suggests that hackers running the attack were keeping count of the number of sites they had compromised, in order to work out how much they stand to get paid.

Further analysis of the tool by security researchers at the SANS Institute's Internet Storm Centre (ISC) is ongoing. The tool came to their attention via a tip-off from Dr Neal Krawetz. Although first thought to be linked to the compromise of thousands of "mom and pop" websites (uncovered by security researcher Mary Landesman, of ScanSafe) it now seems to be related to a different attack.

The constant, changing flux of the malicious JavaScript served up by compromised sites made initial analysis difficult. With the benefit of the hacker tool used to pull off the attack this all becomes much clearer, much like it was easier for scientists to unravel a cure for the mystery pandemic that blighted mankind in the Twelve Monkies after they obtained a sample of the pure source.

"The nice thing about this is that we finally managed to confirm that it is SQL Injection that was used in those attacks. The tool has more functionality that we still have to analyze but this is the main purpose," writes ISC handler Bojan Zdrnja.

Website owners ought to use the discovery as a wake up call on the need to ensure that their web applications are secure, he added. ®

Updated

ScanSafe has been in touch to say that the executable analysed by the ISC is not the source of the mass infection it warned about in January.

"The tool that SANS referenced is not new. It has been known to researchers for some time. There are dozens of these types of tools. These are used for SQL injection attacks and it was not used in the widespread website compromises Mary initially reported on in January (the one that affected thousands of small "mom and pop" sites)," it said.

More about

TIP US OFF

Send us news


Other stories you might like