Wrath of the Slammer worm brings into question a controversial, but common, practice among security researchers.
The security researcher whose code was used as a template for the Slammer worm is now reconsidering his position on releasing sample exploit code.
David Litchfield, co-founder of Next Generation Security Software Ltd., in Surrey, England, said in a post to the BugTraq mailing list Wednesday that hes concerned that some future code of his could be used in a worm that does far more damage than Slammer.
"We often forget that our actions online can have very real consequences in real life," Litchfield wrote. "The next big worm could take out enough critical machines that people are killed. A massive failure of the emergency services computers such as 911/999 could result in someones deathand I dont want to feel that Ive contributed to that."
Now, Litchfield says he will likely decide on a case-by-case basis whether to release exploit code along with his vulnerability advisories. He said he has received more than 100 replies to his BugTraq post, all of which have asked him to continue publishing sample code.
"So it would seem the general consensus is to continue in the current fashion as most have pointed out that the bad guys will know this stuff any way and by posting proof-of-concept code were at least leveling the playing field," Lichfield said. "So this further adds to my indecision. Ill probably end up doing it on a case-by-case basis. Where I think [proof-of-concept code] is needed Ill probably provide itthough after a grace period."
Litchfield last summer discovered the vulnerability in Microsoft Corp.s SQL Server 2000 software that the Slammer worm exploits. At the time he published his advisory on the issue, he also released some sample code showing how the flaw could be exploited. This is a controversial, but common, practice among security researchers. Such code is often included as a proof-of-concept to enable administrators to test their machines for the vulnerability.
But, many software vendors and critics say that publishing sample code leads to quicker exploitation of vulnerabilities by doing some of the crackers work for them. However, the SQL flawand Litchfields codehad been public for six months by the time Slammer hit the Internet last Saturday. And Litchfield said its clear that whoever wrote the worm didnt need much help from him.
"It also becomes apparent that whoever authored the worm knew how to write buffer overflow exploits and would have been capable of doing this without using my shellcode as a template," he wrote. "Having access to my code probably saved them around 20 minutes or so."