The Code Red worm has snarled traffic on the Internet and in other networks by infiltrating hundreds of thousands of servers. How did it accomplish that? Here’s an explanation from Rick Fleming, vice president of security operations at Digital Defense Inc. in San Antonio, Texas: “The Code Red worm works against Microsoft IIS servers only and exploits a buffer overflow condition, giving the hacker the ability to run commands at the system user level. A buffer overflow is a specific type of attack where the hacker inserts more information into the “input” field of a form than the receiving program is expecting. The result is that the buffer, or storage area, in the program receiving the information overflows with information, overwriting part of the program stored in memory. If the hacker can overwrite the correct portion of the program in memory using this technique, then the hacker can cause the program in memory to execute the hacker’s program instead. “Since the attack occurs against the Web server, and most firewalls that protect Web servers have to allow ports 80 and 443 in for the Web server to work properly, a firewall will do nothing to stop Code Red. The most effective way to stop Code Red is to make sure that your IIS servers are patched with the latest software updates. Given the high frequency of IIS servers being involved in computer break-ins, I suggest that users either use a different type server or perform patch update checks weekly.”