Synopsis: Low: kvm security and bug fix update Issue date: 2010-12-20 CVE Names: CVE-2010-3881 It was found that some structure padding and reserved fields in certain data structures in QEMU-KVM were not initialized properly before being copied to user-space. A privileged host user with access to "/dev/kvm" could use this flaw to leak kernel stack memory to user-space. (CVE-2010-3881) This update also fixes the following bugs: * The 'kvm_amd' kernel module did not initialize the TSC (Time Stamp Counter) offset in the VMCB (Virtual Machine Control Block) correctly. After a vCPU (virtual CPU) has been created, the TSC offset in the VMCB should have a negative value so that the virtual machine will see TSC values starting at zero. However, the TSC offset was set to zero and therefore the virtual machine saw the same TSC value as the host. With this update, the TSC offset has been updated to show the correct values. (BZ#656984) * Setting the boot settings of a virtual machine to, firstly, boot from PXE and, secondly, to boot from the hard drive would result in a PXE boot loop, that is, the virtual machine would not continue to boot from the hard drive if the PXE boot failed. This was caused by a flaw in the 'bochs-bios' (part of KVM) code. With this update, after a virtual machine tries to boot from PXE and fails, it continues to boot from a hard drive if there is one present. (BZ#659850) * If a 64-bit Scientific Linux 5.5 virtual machine was migrated to another host with a different CPU clock speed, the clock of that virtual machine would consistently lose or gain time (approximately half a second for every second the host is running). On machines that do not use the kvm clock, the network time protocol daemon (ntpd) could correct the time drifts caused by migration. However, using the pvclock caused the time to change consistently. This was due to flaws in the save/load functions of pvclock. With this update, the issue has been fixed and migrating a virtual machine no longer causes time drift. (BZ#660239) The following procedure must be performed before this update will take effect: 1) Stop all KVM guest virtual machines. 2) Either reboot the hypervisor machine or, as the root user, remove (using "modprobe -r [module]") and reload (using "modprobe [module]") all of the following modules which are currently running (determined using "lsmod"): kvm, ksm, kvm-intel or kvm-amd. 3) Restart the KVM guest virtual machines. SL 5.x SRPMS: kvm-83-164.el5_5.30.src.rpm x86_64: kmod-kvm-83-164.el5_5.30.x86_64.rpm kvm-83-164.el5_5.30.x86_64.rpm kvm-qemu-img-83-164.el5_5.30.x86_64.rpm kvm-tools-83-164.el5_5.30.x86_64.rpm -Connie Sieh -Troy Dawson