DDoS Attacks Hit MySQL Servers

Tuesday, November 3, 2015 @ 06:11 PM gHale

A new Trojan is infecting MySQL servers to conduct distributed denial-of-service (DDoS) attacks.

The Chikdos malware, first documented by Poland’s CERT, has been around since 2013. The malware can hijack Linux and Windows machines and use them to launch DDoS attacks.

New Types of DDoS Attacks
More NTP Holes Fixed
Malware Growing by the Minute
Malware Masquerades as Chrome

The latest versions of the threat are not much different compared to the early variants analyzed by CERT Poland, said researchers at Symantec.

In the attacks observed by the security firm, the attackers have been targeting MySQL servers, most likely because they are widespread and they have large bandwidth that can be helpful for DDoS attacks.

The attacks against MySQL servers start with a malicious user-defined function that acts as a downloader (Downloader.Chikdos) for the actual malware, detected by Symantec as Trojan.Chikdos.A.

User-defined functions end up stored in the file system and they extend the functionality of a MySQL server. These functions usually end up installed by malicious actors on MySQL servers via SQL injection attacks. In this case, researchers said it’s unclear if the attackers have used automated scanning or a worm to compromise the servers and install the user-defined function.

Once executed, the user-defined function downloads the malware from hardcoded URLs (compromised websites) onto the infected server. In some cases, the downloader also adds a new user account to the system.

In the campaign analyzed by Symantec, a quarter of the compromised MySQL servers are in India, followed by China, Brazil, the Netherlands, the U.S., South Korea, Mexico, Canada and Italy. The attacks observed by the security firm aimed at an IP address in China and a hosting provider in the United States.

“To protect against these types of attacks, SQL servers should not be run with administrator privileges where possible. Applications that use the SQL server should be patched regularly and follow good programming practices to mitigate SQL injection vulnerabilities. Check for the presence of new user accounts and ensure that remote access services are configured securely,” Symantec said.