| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| Buffer overflow in the lprm command in the lprold lpr package on SuSE 7.1 through 7.3, OpenBSD 3.2 and earlier, and possibly other operating systems, allows local users to gain root privileges via long command line arguments such as (1) request ID or (2) user name. |
| The arplookup function in FreeBSD 5.1 and earlier, Mac OS X before 10.2.8, and possibly other BSD-based systems, allows remote attackers on a local subnet to cause a denial of service (resource starvation and panic) via a flood of spoofed ARP requests. |
| FreeBSD 4.6 to 4.11 and 5.x to 5.4 uses insecure default permissions for the /dev/iir device, which allows local users to execute restricted ioctl calls to read or modify data on hardware that is controlled by the iir driver. |
| The implementation of SYN cookies (syncookies) in FreeBSD 4.5 through 5.0-RELEASE-p3 uses only 32-bit internal keys when generating syncookies, which makes it easier for remote attackers to conduct brute force ISN guessing attacks and spoof legitimate traffic. |
| Integer overflow in the f_count counter in FreeBSD before 4.2 through 5.0 allows local users to cause a denial of service (crash) and possibly execute arbitrary code via multiple calls to (1) fpathconf and (2) lseek, which do not properly decrement f_count through a call to fdrop. |
| mksnap_ffs in FreeBSD 5.1 and 5.2 only sets the snapshot flag when creating a snapshot for a file system, which causes default values for other flags to be used, possibly disabling security-critical settings and allowing a local user to bypass intended access restrictions. |
| The jail system call in FreeBSD 4.x before 4.10-RELEASE does not verify that an attempt to manipulate routing tables originated from a non-jailed process, which could allow local users to modify the routing table. |
| The jail_attach system call in FreeBSD 5.1 and 5.2 changes the directory of a calling process even if the process doesn't have permission to change directory, which allows local users to gain read/write privileges to files and directories within another jail. |
| FreeBSD 3.2 and possibly other versions allows a local user to cause a denial of service (panic) with a large number accesses of an NFS v3 mounted directory from a large number of processes. |
| cpio on FreeBSD 2.1.0, Debian GNU/Linux 3.0, and possibly other operating systems, uses a 0 umask when creating files using the -O (archive) or -F options, which creates the files with mode 0666 and allows local users to read or overwrite those files. |
| Certain "programming errors" in the msync system call for FreeBSD 5.2.1 and earlier, and 4.10 and earlier, do not properly handle the MS_INVALIDATE operation, which leads to cache consistency problems that allow a local user to prevent certain changes to files from being committed to disk. |
| In some cases, the ktrace facility will log the contents of kernel structures to userspace. In one such case, ktrace dumps a variable-sized sockaddr to userspace. There, the full sockaddr is copied, even when it is shorter than the full size. This can result in up to 14 uninitialized bytes of kernel memory being copied out to userspace.
It is possible for an unprivileged userspace program to leak 14 bytes of a kernel heap allocation to userspace. |
| The command ctl_persistent_reserve_out allows the caller to specify an arbitrary size which will be passed to the kernel's memory allocator. |
| The fetch(3) library uses environment variables for passing certain information, including the revocation file pathname. The environment variable name used by fetch(1) to pass the filename to the library was incorrect, in effect ignoring the option.
Fetch would still connect to a host presenting a certificate included in the revocation file passed to the --crl option. |
| When etcupdate encounters conflicts while merging files, it saves a version containing conflict markers in /var/db/etcupdate/conflicts. This version does not preserve the mode of the input file, and is world-readable. This applies to files that would normally have restricted visibility, such as /etc/master.passwd.
An unprivileged local user may be able to read encrypted root and user passwords from the temporary master.passwd file created in /var/db/etcupdate/conflicts. This is possible only when conflicts within the password file arise during an update, and the unprotected file is deleted when conflicts are resolved. |
| The NVMe driver queue processing is vulernable to guest-induced infinite loops. |
| The virtio_vq_recordon function is subject to a time-of-check to time-of-use (TOCTOU) race condition. |
| A missing null-termination character in the last element of an nvlist array string can lead to writing outside the allocated buffer. |
| In ICMPv6 Neighbor Discovery (ND), the ID is always 0. When pf is configured to allow ND and block incoming Echo Requests, a crafted Echo Request packet after a Neighbor Solicitation (NS) can trigger an Echo Reply. The packet has to come from the same host as the NS and have a zero as identifier to match the state created by the Neighbor Discovery and allow replies to be generated.
ICMPv6 packets with identifier value of zero bypass firewall rules written on the assumption that the incoming packets are going to create a state in the state table. |
| On 64-bit systems, the implementation of VOP_VPTOFH() in the cd9660, tarfs and ext2fs filesystems overflows the destination FID buffer by 4 bytes, a stack buffer overflow.
A NFS server that exports a cd9660, tarfs, or ext2fs file system can be made to panic by mounting and accessing the export with an NFS client. Further exploitation (e.g., bypassing file permission checking or remote kernel code execution) is potentially possible, though this has not been demonstrated. In particular, release kernels are compiled with stack protection enabled, and some instances of the overflow are caught by this mechanism, causing a panic. |
