A vulnerability in filelock versions 3.30.0 through 4.2.0 allows a local user with write access to a shared directory to swap a symbolic link in place of a lock coordination directory during a narrow timing window. This causes the lock user's process to create claim files in a directory chosen by the attacker, potentially disrupting other locks by causing them to time out until manually force-broken. The issue requires the attacker to rename the .filelock directory between an lstat check and subsequent operations, and is limited by directory permissions and sticky bit settings.
Virtualenv's app-data seeder uses shutil.copy to copy cached files into new environments, preserving the original file permissions from the cache. If the cache was created under a permissive umask (e.g., 000), files copied into later environments can retain overly broad permissions (e.g., mode 0666), potentially allowing other local users to modify these files. This issue affects environments where the cache is private but the new environment is accessible by other local accounts. A fix has been introduced to enforce stricter file permissions when copying from the cache.
The ReadWriteLock and AsyncReadWriteLock classes in conda-lock resolve lock file paths by following symlinks and then open these paths with SQLite without preventing symlink traversal. This allows a local attacker with write access to the lock directory to plant symlinks that cause the victim process to create or lock files at unintended locations, potentially breaking mutual exclusion and causing resource corruption.
A vulnerability in the filelock library versions 3.22.0 through 4.2.0 allows a local user with write access to a lock marker file (but without permission to remove it) to forge the process ID in the marker. This causes another process running as the lock owner or root to mistakenly consider the lock stale, remove the live marker, and acquire the lock simultaneously. This results in two processes concurrently entering a critical section without error, potentially corrupting integrity of the protected resource. The issue arises when the marker file is created with group or other write permissions, typically due to an explicit mode or permissive umask settings in a sticky or non-removable directory. The vulnerability has a CVSS 3.1 score of 4.7 (Medium) and requires local access with low privileges and a high attack complexity. A fix is available in version 3.27.1_2.
CVE-2026-92092 is a vulnerability in the openshift-ansible installer container for Red Hat OpenShift Container Platform 4.14. The installer’s run script downloads and executes remote inventory and dynamic scripts without verifying their integrity. An attacker with the ability to perform a Man-in-the-Middle (MITM) attack or compromise the HTTP server hosting these scripts could substitute malicious executables. These executables would run with the installer's user ID and full cluster-provisioning privileges, enabling arbitrary code execution. The risk is limited to environments that use remote inventory sources via environment variables and do not use secure HTTPS URLs. Local inventory usage or isolated networks are not affected. The vulnerability is rated high severity with a CVSS score of 7.4.
Two serious vulnerabilities were found in the Rust zip crate, a widely used library for handling ZIP files. The first (RUSTSEC-2026-0367) allows privilege escalation by preserving dangerous file permission bits (setuid, setgid, sticky bit) when extracting archives, potentially enabling attackers to plant setuid-root files. The second (CVE-2026-115905) is a critical remote code execution flaw caused by inconsistent handling of file names between the central directory and local headers in ZIP archives, allowing arbitrary file overwrite and execution. Both vulnerabilities are fixed in version 9.0.5 of the zip crate. The advisory emphasizes that patching alone is insufficient and advocates architectural security using deny-by-default sandboxes to contain untrusted code and limit the impact of such vulnerabilities.
Microsoft has been named a Leader in the 2026 IDC MarketScape for Worldwide Modern Endpoint Security for Enterprises. The recognition highlights Microsoft Defender's autonomous attack disruption and predictive shielding capabilities that contain ransomware and identity attacks within minutes. The platform integrates AI-driven detection and response across endpoints, identities, email, and cloud, enabling near real-time containment of threats. The solution includes agentic AI through Security Copilot, improving alert accuracy and reducing analyst workload. Microsoft Defender's unified portal correlates signals from multiple sources into single incidents, facilitating rapid and effective response. The platform also supports securing local AI agents used by employees, providing visibility and control over their activity. This integrated approach addresses the increasing speed and complexity of ransomware and AI-driven attacks observed between 2025 and 2026.
A vulnerability in dnsmasq's DHCPv6 client hostname parser allows reading one byte past the end of a crafted FQDN option. This out-of-bounds read occurs because the parser checks the next label length without confirming the pointer remains inside the option. The flaw can be triggered by an attacker on the local network with a single DHCPv6 packet. No crash or data leak has been demonstrated, and the extra byte is not returned or logged.
A vulnerability in dnsmasq occurs when the DHCP leasequery feature is enabled without restricting allowed source addresses. This flaw allows any host on the local network to request DHCP lease records, revealing assigned IP addresses, MAC addresses, hostnames, and lease metadata. The leasequery feature is disabled by default and must be explicitly enabled by an administrator. Red Hat rates this vulnerability as low impact because the exposed data is not highly sensitive and the feature is opt-in.