Just in time to counter security taints from last week's MySpooler worm, which spread via weak MySQL passwords on Windows installations, MySQL on Friday got a clean bill of health from code analysis firm Coverity Inc.
The five Stanford University researchers at Coverity, who analyzed the security of the Linux kernel over a period of four years, this month are planning to release an analysis of the security and quality of MySQL code that found the database to have an "excellent" bug density. ADVERTISEMENT
Coverity did the analysis at the request of MySQL AB, the company that markets and develops the code for the open-source database under a dual-licensing structure.
Coverity researchers analyzed MySQL Version 4.1.8 in January. The types of defects discovered were crash-causing defects, performance degradation and security vulnerabilities.
Out of 425,000 lines of code analyzed, Coverity identified 97 bugs, which included Deadcode, or unused code due to logic flaws, which can lead to improper system function; forward null, which can cause system crash; negative returns, static overrun and overrun dynamic, all of which can cause data corruption, possible crash or possible malicious attack; resource leak; reverse null; uninitialized variable; and unused value.
The company has used the same code-checking technology on source code from customers including Oracle Corp. and Veritas Software Corp., among others.
PointerIn the wake of the MySpooler worm, users criticized MySQL for not hardening the database. Read more here.
For comparison's sake, Coverity found more than 1,000 defects in the Linux kernel Version 2.4.1 in 2001, at a time when the source code contained 1.6 million lines of code.
Coverity CEO Seth Hallem said the relative cleanliness of MySQL code is likely attributable to a few things: First, it is shepherded by MySQL AB, the Uppsala, Sweden-based company that markets and develops the database under a dual-licensing structure. "They're putting their necks on the line," he said. "They're certainly more interested than a project with less industry penetration or less of a profit motive."
No comments:
Post a Comment