Security update for Linux kernel
(Last modified: 23AUG2005)
solutions Security update for Linux kernel SuSE Linux Maintenance Web (17b472d8726ee7e9fa73b9a5cdb787ff)
Product(s): SUSE CORE 9 for IBM zSeries 64bit
- CAN-2005-2457: A problem in decompression of files on "zisofs" filesystem was fixed.
- CAN-2005-2458: A potential buffer overflow in the zlib decompression handling in the kernel was fixed.
- CAN-2005-2459: Some return codes in zlib decoding were fixed which could have led to an attacker crashing the kernel.
- CAN-2005-2555: Only CAP_NET_ADMIN is now allowed load socket policies.
- CAN-2005-0916: Fixed a possible crash in the AIO hugepage handling of PowerPC64.
- CAN-2005-2456: Fixed a potential overflow caused by missing boundary checks of sock->sk_policy in net/xfrm/.
- x86_64: A previous fix for a denial of service attack with compat 32bit mode programs was too strict and could crash the kernel. (The earlier fix had the Mitre ID CAN-2005-1765.)
- s390: Fixed /sys/ permissions where a user could change machine states, including powering down or up partitions.
- The reported process start times sometimes were incorrect.
- The OCFS2 filesystem was updated to version 1.0.2.
- A potential deadlock in cpuset handling was fixed.
- Fixed a potential crash on startup of the tg3 network driver.
- Avoid high IRQ latencies in the VM handling.
- rpm/post.sh was fixed so that initrd.previous is preserved again.
- A problem in the handling of the tape ioctl MTIOCPOS was fixed.
- Make the OOM process killer send SIGTERM first instead of SIGKILL.
- Fixed a netfilter connection track return code mismatch.
- Fixed a typo in the ipt_TTL netfilter module.
- XEN was updated to version 2.0.6b.
- Allow rsize/wsize values less than 4096 for NFS mounts.
- A data corruption problem within the reiserfs filesystem in the handling of writing to mmaped regions after close of the filedescriptor was fixed.
- Problem-ID: 16923 - kernel: internal return values returned to userspace.
- Problem-ID: 14978 - zfcp: I/O errors on data underruns during Flash Copy on ESS800
- Problem-ID: 16967 - kernel: incorrect posix fadvise values.
- Problem-ID: 17349 - s390dbf: Inconsistent output of debug areas. Wrong timestamps of debug entries.
- Problem-ID: 17420 - ctcmpc: Missing storage free in exception case.
- Problem-ID: 17452 - cio: I/O blocked after cable pull/plug with all path gone.
- Problem-ID: 17438 - qeth: Disable IPV6 for hipersockets.
- Problem-ID: 17454 - qeth: Hipersockets do not work with enabled fakell option.
- Problem-ID: 17120 - qeth-channel bonding ping causes skb resource problems
- Problem-ID: 16811 - qeth: recovery problem after vary off/on
- Problem-ID: 17067 - A race condition and a busy wait performance problem caused by behaviour of the transmit function
- add patches.arch/s390-spinretry.diff Retryable spinlocks for S/390 (#100722 - LTC17469).
- patches.arch/s390-ctc-busy-wait.patch: Fix busy wait in CTC driver (97075 - LTC17067).
- patches.arch/s390-qeth-recover.patch Properly recover qeth devices after vary on (94842 - LTC16811).
rpm -Fvh kernel-s390x.rpm kernel-source.rpm kernel-syms.rpm zipl
Finally, reboot the system with
shutdown -r now
to load the new kernel (replace "now" with the appropriate amount of time to allow local users to cleanly log out, for example "+5" for five minutes.)
Download Source Packages
Download the source code of the patches for maintained products.
The Origin of this information may be internal or external to Novell. Novell makes all reasonable efforts to verify this information. However, the information provided in this document is for your information only. Novell makes no explicit or implied claims to the validity of this information.
Any trademarks referenced in this document are the property of their respective owners. Consult your product manuals for complete trademark information.