We will be scheduling an important kernel security and bug fix update for all our CentOS 5 and 6 servers this weekend.

This would take place as far as possible during the off-peak hours. A reboot is required to complete the upgrade. The downtime should not exceed 30 minutes and it will be minimize as much as possible.


This update is scheduled as follows:

Date: 18 Jan 2014 (Saturday) to 20 Jan 2014 (Monday)
Time: Between 2AM and 8AM EST

Details of CentOS 5 security update

* An information leak flaw was found in the way the Linux kernel's device mapper subsystem, under certain conditions, interpreted data written to snapshot block devices. An attacker could use this flaw to read data from disk blocks in free space, which are normally inaccessible. (CVE-2013-4299, Moderate)

This update also fixes the following bugs:

* A previous fix to the kernel did not contain a memory barrier in the percpu_up_write() function. Consequently, under certain circumstances, a race condition could occur, leading to memory corruption and a subsequent kernel panic. This update introduces a new memory barrier pair, light_mb() and heavy_mb(), for per-CPU basis read and write semaphores (percpu-rw-semaphores) ensuring that the race condition can no longer occur. In addition, the read path performance of "percpu-rw-semaphores" has been improved. (BZ#884735)

* Due to several related bugs in the be2net driver, the driver did not handle firmware manipulation of the network cards using the Emulex XE201 I/O controller properly. As a consequence, these NICs could not recover from an error successfully. A series of patches has been applied that fix the initialization sequence, and firmware download and activation for the XE201 controller. Error recovery now works as expected for the be2net NICs using the Emulex XE201 I/O controller. (BZ#1019892)

* A bug in the be2net driver could cause packet corruption when handling VLAN-tagged packets with no assigned VLAN group. This happened because the be2net driver called a function responsible for VLAN tag reinsertion in a wrong order in the code. The code has been restructured and the be2net driver now calls the __vlan_put_tag() function correctly, thus avoiding the packet corruption. (BZ#1019893)

* A previous patch to the kernel introduced the "VLAN tag re-insertion" workaround to resolve a problem with incorrectly handled VLAN-tagged packets with no assigned VLAN group while the be2net driver was in promiscuous mode. However, this solution led to packet corruption and a subsequent kernel oops if such a processed packet was a GRO packet. Therefore, a patch has been applied to restrict VLAN tag re-insertion only to non-GRO packets. The be2net driver now processes VLAN-tagged packets with no assigned VLAN group correctly in this situation. (BZ#1023347)

Details of CentOS 6 security update:

None.

This update fixes the following bugs:

* Previously, the bonding driver did not support TCP segmentation offloading (TSO). As a consequence, the performance decreased significantly, and the "ethtool -k bondX" command reported that the TSO was disabled, although the slave device had the TSO enabled. The following message appeared in the kernel log: "Dropping TSO features since no CSUM". The problem affected also bridges and loopback interfaces. The underlying source code has been modified to ensure that the bonding module does not drop TSO features for NO_CSUM devices. As a result, the aforementioned performance problem no longer occurs. (BZ#1037812)

* There was an error in the tag insertion logic, and bonding handled cases when a slave device did not have a hardware VLAN acceleration. As a consequence, the network packets were tagged twice through slave devices without hardware VLAN tag insertion, and two Solarflare network cards using bonding over a VLAN did not work properly. This update removes the redundant VLAN tag insertion logic, and the unwanted behavior no longer occurs. (BZ#1039562)

Venerdì, Gennaio 17, 2014

« Indietro

Powered by WHMCompleteSolution