Security

• Lurker's mechanism for specifying configuration files was vulnerable to being overridden. As lurker includes sections of unparsed config files in its output, an attacker could manipulate lurker into reading any file readable by the www-data user. (CVE-2006-1062)
• It is possible for a remote attacker to create or overwrite files in any writable directory that is named "mbox". (CVE-2006-1063)
• Missing input sanitizing allows an attacker to inject arbitrary web script or HTML. (CVE-2006-1064)

Ludwig Nussel discovered an integer overflow bug in the way gdk-pixbuf processes XPM images. An attacker could create a carefully crafted XPM file in such a way that it could cause an application linked with gdk-pixbuf to execute arbitrary code or crash when the file was opened by a victim.

Ludwig Nussel also discovered an infinite-loop denial of service bug in the way gdk-pixbuf processes XPM images. An attacker could create a carefully crafted XPM file in such a way that it could cause an application linked with gdk-pixbuf to stop responding when the file was opened by a victim.

Tetsuo Handa discovered a local Denial of Service vulnerability in the udp_v6_get_port() function. On computers which use IPv6, a local attacker could exploit this to trigger an infinite loop in the kernel. (CVE-2005-2973)

Harald Welte discovered a Denial of Service vulnerability in the USB devio driver. A local attacker could exploit this by sending an "USB Request Block" (URB) and terminating the sending process before the arrival of the answer, which left an invalid pointer and caused a kernel crash. (CVE-2005-3055)

Pavel Roskin discovered an information leak in the Orinoco wireless card driver. When increasing the buffer length for storing data, the buffer was not padded with zeros, which exposed a random part of the system memory to the user. (CVE-2005-3180)

A resource leak has been discovered in the handling of POSIX timers in the exec() function. This could be exploited to a Denial of Service attack by a group of local users. (CVE-2005-3271)

Stephen Hemminger discovered a weakness in the network bridge driver. Packets which had already been dropped by the packet filter could poison the forwarding table, which could be exploited to make the bridge forward spoofed packages. (CVE-2005-3272)

David S. Miller discovered a buffer overflow in the rose_rt_ioctl() function. By calling the function with a large "ngidis" argument, a local attacker could cause a kernel crash. (CVE-2005-3273)

Neil Horman discovered a race condition in the connection timer handling. This allowed a local attacker to set up an expiration handler which modified the connection list while the list still being traversed, which could result in a kernel crash. This vulnerability only affects multiprocessor (SMP) systems. (CVE-2005-3274)

Patrick McHardy noticed a logic error in the network address translation (NAT) connection tracker. A remote attacker could exploit this by causing two packets for the same protocol to be NATed at the same time, which resulted in a kernel crash. (CVE-2005-3275)

Paolo Giarrusso discovered an information leak in the sys_get_thread_area(). The returned structure was not properly cleared, which exposed a small amount of kernel memory to userspace programs. This could possibly expose confidential data. (CVE-2005-3276)

• A race condition in the 2.6 kernel could allow a local user to cause a DoS by triggering a core dump in one thread while another thread has a pending SIGSTOP (CVE-2005-3527).
• The ptrace functionality in 2.6 kernels prior to 2.6.14.2, using CLONE_THREAD, does not use the thread group ID to check whether it is attaching to itself, which could allow local users to cause a DoS (CVE-2005-3783).
• The auto-reap child process in 2.6 kernels prior to 2.6.15 include processes with ptrace attached, which leads to a dangling ptrace reference and allows local users to cause a crash (CVE-2005-3784).
• A locking problem in the POSIX timer cleanup handling on exit on kernels 2.6.10 to 2.6.14 when running on SMP systems, allows a local user to cause a deadlock involving process CPU timers (CVE-2005-3805).
• The IPv6 flowlabel handling code in 2.4 and 2.6 kernels prior to 2.4.32 and 2.6.14 modifies the wrong variable in certain circumstances, which allows local users to corrupt kernel memory or cause a crash by triggering a free of non-allocated memory (CVE-2005-3806).
• An integer overflow in 2.6.14 and earlier could allow a local user to cause a hang via 64-bit mmap calls that are not properly handled on a 32-bit system (CVE-2005-3808).