Fourth new Linux kernel milestone of 2017 is now available, providing improved performance and security features including Kernel Transport Layer Security
While most Americans were enjoying the Labor Day weekend, Linux creator Linus Torvalds was busy releasing the Linux 4.13 kernel on Sept. 3. Linux 4.13 is the fourth new Linux kernel released in 2017 and follows Linux 4.12, which debuted in July.
The Linux 4.13 kernel is noteworthy for a number of reasons, including multiple security-related enhancements and some health issues that Torvalds dealt with during the development cycle, which included seven release candidates.
“The other excitement this week was purely personal, consisting of seven hours of pure agony due to a kidney stone,” Torvalds wrote. “I’m all good, but it sure _felt_ a lot longer than seven hours, and I don’t even want to imagine what it is for people that have had the experience drag out for longer.”
Among the security-related changes in the Linux 4.13 kernel is one that Torvalds referred to as a generic protocol issue.
“The change in question is simply changing the default cifs behavior: instead of defaulting to SMB 1.0 (which you really should not use: just google for ‘stop using SMB1’ or similar), the default cifs mount now defaults to a rather more modern SMB 3.0,” Torvalds wrote.
The Common Internet File System (CIFS) and the Server Message Block (SMB) protocol enable cross-platform file and folder sharing between different Windows and Linux systems. SMB version 1 has been considered to be insecure for several years, with multiple vendors including Microsoft warning users since at least 2016 not to use it. SMB-related vulnerabilities have also been a prominent component of several recent high-profile ransomware attacks, including both WannaCry in May and NotPetya in June
According to Torvalds, most Linux users should not notice the change to the newer SMB 3.0 protocol as the default for CIFS. He noted that for those who do notice the change, they should still move away from SMB version 1 to a newer version.
“Because let’s face it, SMB1 is just bad, bad, bad,” Torvalds wrote.
Facebook engineer Dave Watson originally proposed the idea of KTLS as a way to accelerate TLS performance at scale.
“In kernel implementations provide new opportunities for optimization of TLS,” Watson wrote in a research paper. “Our implementation saves up to 7 percent CPU copy overhead and up to 10 percent latency improvements when combined with the Kernel Connection Multiplexor (KCM).”
Originally published on eWeek