High7.8KEVEPSS 1.9%Check Point ResearchCVE-2026-123039
Starting October 30, 2026, attackers exploited unpatched Check Point Security Gateways via Mobile Access portals using PHP2Shell for initial access, followed by privilege escalation through CVE-2026-123039. Initial attacks targeted Israeli organizations with destructive activity, while a broader second wave targeted multiple countries without destructive actions. Attackers deployed PHP webshells, the persistent TSGA backdoor with TSProxy, and later pivoted to VMware ESXi hosts to deploy a destructive wiper and leave politically themed notes. The malware tools are linked to the PhiliKit framework, previously associated with China nexus threat actors, though attribution remains uncertain. The campaign exploited multiple vulnerabilities and established persistent, covert access on compromised systems.
Two security vulnerabilities affecting the Cockpit web-based server administration tool in Red Hat Enterprise Linux 9.8 Extended Update Support have been addressed. One vulnerability (CVE-2026-91151) allows local privilege escalation due to insecure symbolic link handling in the public key addition helper. The other (CVE-2026-91150) is a denial of service caused by unbounded fragmented WebSocket messages. Red Hat has released updates to fix these issues in the Cockpit packages for multiple architectures.
In Progress® Telerik® Reporting prior to version 20.3.26.1110, improper restriction of XML external entity references in the SVG processing engine allows an authenticated attacker with low privileges who can supply SVG content to a report to read local or UNC files, or trigger server-side network requests, from the report rendering host. External XML entities in SVG content were resolved without applying the configured strict resource resolver and report path validation. Disclosed content is limited by the permissions of the rendering process and by what the rendered report output exposes.
An Incorrect Authorization vulnerability exists in TÜBİTAK BİLGEM Software Technologies Research Institute Pardus About prior to version 1.2.5, which allows privilege escalation. The vulnerability has a high severity with a CVSS score of 7.8. It affects local attack vectors with low complexity and no user interaction required, potentially leading to full confidentiality, integrity, and availability impact.
Microsoft's FORGE Lab conducted an AI-assisted security scan of Linux kernel subsystems, resulting in 59 public upstream vulnerability reports and 33 CVEs. Among these, eight CVEs had confirmed local root privilege-escalation exploits. The vulnerabilities, many of which had existed for years (average age 8.1 years), include use-after-free bugs, refcount underflows, and buffer overflows in critical kernel components such as networking and security modules. Fixes for 38 reports have been merged into the Linux mainline kernel. The research highlights the complexity and scale of vulnerability discovery in the Linux kernel and the role of AI in accelerating this process.
A vulnerability in BusyBox's tar utility allows improper resolution of deferred hardlink targets through symbolic links pointing outside the intended extraction directory. This flaw can be exploited by an attacker providing a specially crafted archive to overwrite arbitrary files on the system during extraction. The issue requires local user interaction and high attack complexity, limiting exposure primarily to constrained environments using BusyBox for extraction. Red Hat notes minimal exposure in common deployments and currently has no mitigation or fix meeting their criteria.
CVE-2026-110992 is a path traversal vulnerability in Grafana OSS that affects versions 13.0.0 through 13.3.0 (with some gaps). A user with the Editor role who can create Git Sync provisioning jobs can exploit this flaw by referencing a resource with a source path containing '../' sequences. This allows deletion or movement of files outside the configured repository path on the local filesystem, potentially impacting files on the Grafana host.
Red Hat has issued a security advisory for gvfs, the GNOME Desktop Virtual File System layer, addressing two vulnerabilities: a heap-based buffer overflow in the SFTP read_reply() function (CVE-2026-84268) and a race condition in gvfs-admin socket ownership that permits local root escalation (CVE-2026-88924). These issues affect Red Hat Enterprise Linux 10 across multiple architectures. An update is available to fix these vulnerabilities.
A local user can exploit a flaw in the multipath-tools package by sending a specially crafted command with a format string ending in a trailing percent character to the multipath daemon. This triggers an out-of-bounds read, potentially exposing unintended daemon memory and causing information disclosure or crashing the daemon, resulting in denial of service. The vulnerability affects versions prior to 0.9.5-23.oe2403sp4 and has a medium severity rating with a CVSS score of 5.1. A security update fixing this issue is available in version 0.9.5-23.oe2403sp4.
Multiple security vulnerabilities have been identified in Flatpak, a system for building, distributing, and running sandboxed desktop applications on Linux. These include arbitrary code execution, arbitrary write access as root, buffer overflow on 32-bit systems, local privilege escalation, arbitrary file deletion, and arbitrary write in root context via various path traversal and symlink issues. The vulnerabilities affect Red Hat Enterprise Linux 9 and related CodeReady Linux Builder products across multiple architectures. Red Hat has issued an important security advisory with updates to address these issues.
A security update for the Ignition utility in Red Hat Enterprise Linux 9 addresses a vulnerability in the golang crypto/x509 package that could allow man-in-the-middle attacks via untrusted intermediate certificates. Ignition is used during system initialization to configure disks, files, and users based on remote or local configuration sources. The update fixes CVE-2026-111943 to prevent exploitation of this certificate validation flaw.
A vulnerability in Progressive Robot hMailServer versions 6.2.25 through 6.3.7 on Windows affects configurations using a PostgreSQL database with certain SSL modes and root certificate file settings. The issue arises from how the root certificate file name is handled with Windows code pages and UTF-8 encoding, allowing a local attacker to cause hMailServer to trust a malicious certificate authority. This enables interception and modification of database traffic, including the database password. The vulnerability does not affect default configurations and is specific to Windows builds with PostgreSQL or Oracle MySQL client libraries replacing the bundled MariaDB Connector/C.
CVE-2026-126621 is a vulnerability in Progressive Robot Ltd's hMailServer versions 6.0.0-B1 through 6.3.7 on Windows. It involves improper certificate validation due to how OpenSSL handles TLS file names encoded as Latin-1 bytes on Windows. This flaw allows a local attacker with file creation permissions to substitute TLS-related files, potentially enabling unauthorized client certificate authentication, interception and modification of stored messages, access to private keys used by the ACME client, or denial of TLS port startup. The vulnerability requires non-default configurations involving non-ASCII file names and enabled client certificates or object store features. Linux builds are not affected.
CVE-2026-126611 is a vulnerability in Progressive Robot Ltd's hMailServer versions 6.3.4 through 6.3.7 when configured with a MySQL or MariaDB database, TlsMode=verify-full, and no TlsCaFile. The server does not properly verify the database server's certificate issuer or name for connections treated as local by MariaDB Connector/C, allowing a local attacker to impersonate the database server. This can lead to interception of database credentials and modification of traffic between hMailServer and the database. The vulnerability arises because MariaDB Connector/C accepts any valid self-signed certificate or one with its own root without proper verification in these conditions. The default configuration is not affected as it uses TlsMode=prefer, which does not enforce certificate verification.