Recently, CTO of Azul Systems and co-founder Gil Tene reported a very important but little-known Linux kernel patch at GoogleGroups, users and administrators of Linux systems using Intel Haswell architecture should pay special attention to this issue. In particular, users based on Red Hat distributions (including CentOS 6.6 and ScientificLinux6.6) should immediately update the patch. Even Linux running on a virtual machine, if the virtual machine is on a popular cloud platform (such as Azure and Amazon), it may also run on Haswell Machines. patching should be advantageous. Tene describes the defect as follows: The impact of this kernel vulnerability is very simple: in some seemingly impossible situations, user processes will be deadlocked and suspended. Any futex call wait (even if it is correctly awakened) may always be blocked for execution. Just as Thread. park () in Java may be congested all the time, and so on. If you are lucky enough, you will find the soft lockup message in the dmesg log; if you are not so lucky (such as with us ), you will have to spend several months manually investigating problems in the code, and you may have nothing to gain." Tene continues to explain how the code with this defect is executed (eventually a switch block with default missing ). The biggest problem now is that, although the problem code was fixed in January 2014, it was moved back to the Red Hat October 2014 family system around 6.6. Other systems, including SLES, Ubuntu, and Debian, may also be affected. These system fixes are currently inconsistent and may be ignored. RedHat users should use RHEL 6.6.z or an updated version. Tene also points out that different releases have different options for what to put into the kernel, which leads to inconsistent fixes. For example, for RHEL 7.1, "In fact, the upstream 3.10 kernel does not have this bug, but the RHEL7 kernel is not a pure upstream version. Unfortunately, RHEL 7.1 (like RHEL6.6) included the (based on RHEL 7) bug during porting ...... I think other releases may do the same." Tene provides a quick reference list for RHEL-based releases: L RHEL 5 (including CentOS5 and Scientific Linux 5): all versions (including version 5.11) are correct. L RHEL 6 (including CentOS6 and Scientific Linux 6): from 6.0 ~ Version 6.5 is okay. However, version 6.6 is defective, and version 6.6.z is normal. L RHEL 7 (including CentOS7 and Scientific Linux 7): 7.1 is defective. As of January 1, May 13, 2015, there was no 7.x fix. Although there are some disputes over the number of affected systems on Hacker News, it provides some environments to check whether your system needs to be repaired. Get free LAMP Brothers original linux O & M engineer video/detailed linux tutorial, details consulting official website Customer Service: http://www.lampbrother.net/linux/
|