JFFS2, as used on One Laptop Per Child (OLPC) build 542 and possibly other Linux systems, when POSIX ACL support is enabled, does not properly store permissions during (1) inode creation or (2) ACL setting, which might allow local users to access restricted files or directories after a remount of a filesystem, related to "legacy modes" and an inconsistency between dentry permissions and inode permissions.

Red Hat Enterprise Linux (RHEL) 5 ships the rpm for the Advanced Intrusion Detection Environment (AIDE) before 0.13.1 with a database that lacks checksum information, which allows context-dependent attackers to bypass file integrity checks and modify certain files.

xterm, including 192-7.el4 in Red Hat Enterprise Linux and 208-3.1 in Debian GNU/Linux, sets the wrong group ownership of tty devices, which allows local users to write data to other users' terminals.

Untrusted search path vulnerability in the wrapper scripts for the (1) rug, (2) zen-updater, (3) zen-installer, and (4) zen-remover programs on SUSE Linux 10.1 and Enterprise 10 allows local users to gain privileges via modified (a) LD_LIBRARY_PATH and (b) MONO_GAC_PREFIX environment variables.

Multiple untrusted search path vulnerabilities in IBM DB2 UDB 8 before Fixpak 15 and 9.1 before Fixpak 3 allow local users to gain privileges via certain vectors related to (1) DB2 instance or FMP startup on Linux and Solaris; (2) exec of executables while running as root on non-Windows systems, as demonstrated by AIX; and unspecified vectors involving (3) db2licm and (4) db2pd.

The installation script for orarun on SUSE Linux before 20070810 places the oracle user into the disk group, which allows the local oracle user to read or write raw disk partitions.

Unspecified vulnerability in a "core clean" cron job created by the findutils-locate package on SUSE Linux 10.0 and 10.1 and Enterprise Server 9 and 10 before 20070810 allows local users to delete of arbitrary files via unknown vectors.

Linux kernel 2.4.35 and other versions allows local users to send arbitrary signals to a child process that is running at higher privileges by causing a setuid-root parent process to die, which delivers an attacker-controlled parent process death signal (PR_SET_PDEATHSIG).

The (1) aac_cfg_open and (2) aac_compat_ioctl functions in the SCSI layer ioctl path in aacraid in the Linux kernel before 2.6.23-rc2 do not check permissions for ioctls, which might allow local users to cause a denial of service or gain privileges.

The xfer_secondary_pool function in drivers/char/random.c in the Linux kernel 2.4 before 2.4.35 performs reseed operations on only the first few bytes of a buffer, which might make it easier for attackers to predict the output of the random number generator, related to incorrect use of the sizeof operator.

The drm/i915 component in the Linux kernel before 2.6.22.2, when used with i965G and later chipsets, allows local users with access to an X11 session and Direct Rendering Manager (DRM) to write to arbitrary memory locations and gain privileges via a crafted batchbuffer.

The Linux kernel before 2.6.23-rc1 checks the wrong global variable for the CIFS sec mount option, which might allow remote attackers to spoof CIFS network traffic that the client configured for security signatures, as demonstrated by lack of signing despite sec=ntlmv2i in a SetupAndX request.

The default configuration of Centre for Speech Technology Research (CSTR) Festival 1.95 beta (aka 2.0 beta) on Gentoo Linux, SUSE Linux, and possibly other distributions, is run locally with elevated privileges without requiring authentication, which allows local and remote attackers to execute arbitrary commands via the local daemon on port 1314, a different vulnerability than CVE-2001-0956. NOTE: this issue is local in some environments, but remote on others.

NVIDIA drivers (nvidia-drivers) before 1.0.7185, 1.0.9639, and 100.14.11, as used in Gentoo Linux and possibly other distributions, creates /dev/nvidia* device files with insecure permissions, which allows local users to modify video card settings, cause a denial of service (crash or physical video card damage), and obtain sensitive information.

The CUPS service, as used in SUSE Linux before 20070720 and other Linux distributions, allows remote attackers to cause a denial of service via unspecified vectors related to an incomplete fix for CVE-2007-0720 that introduced a different denial of service problem in SSL negotiation.

Stack-based buffer overflow in the random number generator (RNG) implementation in the Linux kernel before 2.6.22 might allow local root users to cause a denial of service or gain privileges by setting the default wakeup threshold to a value greater than the output pool size, which triggers writing random numbers to the stack by the pool transfer function involving "bound check ordering". NOTE: this issue might only cross privilege boundaries in environments that have granular assignment of privileges for root.

Rule Set Based Access Control (RSBAC) before 1.3.5 does not properly use the Linux Kernel Crypto API for the Linux kernel 2.6.x, which allows context-dependent attackers to bypass authentication controls via unspecified vectors, possibly involving User Management password hashing and unchecked function return codes.

The Distributed Lock Manager (DLM) in the cluster manager for Linux kernel 2.6.15 allows remote attackers to cause a denial of service (loss of lock services) by connecting to the DLM port, which probably prevents other processes from accessing the service.

Unspecified vulnerability in HP ServiceGuard for Linux for Red Hat Enterprise Linux (RHEL) 2.1 SG A.11.14.04 through A.11.14.06; RHEL 3.0 SG A.11.16.04 through A.11.16.10; and ServiceGuard Cluster Object Manager B.03.01.02 allows local users to gain privileges via unspecified vectors, a different vulnerability than CVE-2007-0980.

Unspecified vulnerability in Pidgin (formerly Gaim) 2.0.2 for Linux allows remote authenticated users, who are listed in a users list, to execute certain commands via unspecified vectors, aka ZD-00000035. NOTE: this information is based upon a vague advisory by a vulnerability information sales organization that does not coordinate with vendors or release actionable advisories. A CVE has been assigned for tracking purposes, but duplicates with other CVEs are difficult to determine.