| Description: | The kernel packages contain the Linux kernel, the core of any Linux operating system.
Security fixes:
a NULL pointer dereference flaw was found in the Linux kernel NFSv4 implementation. Several of the NFSv4 file locking functions failed to check whether a file had been opened on the server before performing locking operations on it. A local, unprivileged user on a system with an NFSv4 share mounted could possibly use this flaw to cause a kernel panic (denial of service) or escalate their privileges. (CVE-2009-3726, Important)
a flaw was found in the sctp_process_unk_param() function in the Linux kernel Stream Control Transmission Protocol (SCTP) implementation. A remote attacker could send a specially-crafted SCTP packet to an SCTP listening port on a target system, causing a kernel panic (denial of service). (CVE-2010-1173, Important)
a race condition between finding a keyring by name and destroying a freed keyring was found in the Linux kernel key management facility. A local, unprivileged user could use this flaw to cause a kernel panic (denial of service) or escalate their privileges. (CVE-2010-1437, Important)
Red Hat would like to thank Simon Vallet for responsibly reporting CVE-2009-3726; and Jukka Taimisto and Olli Jarva of Codenomicon Ltd, Nokia Siemens Networks, and Wind River on behalf of their customer, for responsibly reporting CVE-2010-1173.
Bug fixes:
RHBA-2007:0791 introduced a regression in the Journaling Block Device (JBD). Under certain circumstances, removing a large file (such as 300 MB or more) did not result in inactive memory being freed, leading to the system having a large amount of inactive memory. Now, the memory is correctly freed. (BZ#589155)
the timer_interrupt() routine did not scale lost real ticks to logical ticks correctly, possibly causing time drift for 64-bit Red Hat Enterprise Linux 4 KVM (Kernel-based Virtual Machine) guests that were booted with the "divider=x" kernel parameter set to a value greater than 1. "warning: many lost ticks" messages may have been logged on the affected guest systems. (BZ#590551)
a bug could have prevented NFSv3 clients from having the most up-to-date file attributes for files on a given NFSv3 file system. In cases where a file type changed, such as if a file was removed and replaced with a directory of the same name, the NFSv3 client may not have noticed this change until stat(2) was called (for example, by running "ls -l"). (BZ#596372)
RHBA-2007:0791 introduced bugs in the Linux kernel PCI-X subsystem. These could have caused a system deadlock on some systems where the BIOS set the default Maximum Memory Read Byte Count (MMRBC) to 4096, and that also use the Intel PRO/1000 Linux driver, e1000. Errors such as "e1000: eth[x]: e1000_clean_tx_irq: Detected Tx Unit Hang" were logged. (BZ#596374)
an out of memory condition in a KVM guest, using the virtio-net network driver and also under heavy network stress, could have resulted in that guest being unable to receive network traffic. Users had to manually remove and re-add the virtio_net module and restart the network service before networking worked as expected. Such memory conditions no longer prevent KVM guests receiving network traffic. (BZ#597310)
when an SFQ qdisc that limited the queue size to two packets was added to a network interface, sending traffic through that interface resulted in a kernel crash. Such a qdisc no longer results in a kernel crash. (BZ#597312)
when an NFS client opened a file with the O_TRUNC flag set, it received a valid stateid, but did not use that stateid to perform the SETATTR call. Such cases were rejected by Red Hat Enterprise Linux 4 NFS servers with an "NFS4ERR_BAD_STATEID" error, possibly preventing some NFS clients from writing files to an NFS file system. (BZ#597314)
Users should upgrade to these updated packages, which contain backported patches to correct these issues. The system must be rebooted for this update to take effect.
|