Deserialization of Untrusted Data

Overview megatron-bridge is a Megatron Bridge: Training Recipes for Megatron-based LLM and VLM models Affected versions of this package are vulnerable to Deserialization of Untrusted Data via the deserialization process. An attacker can execute arbitrary code, escalate privileges, tamper with dat...

Deserialization of Untrusted Data

Overview megatron-bridge is a Megatron Bridge: Training Recipes for Megatron-based LLM and VLM models Affected versions of this package are vulnerable to Deserialization of Untrusted Data via the deserialization process. An attacker can execute arbitrary code, escalate privileges, tamper with dat...

Arbitrary Code Injection

Overview megatron-bridge is a Megatron Bridge: Training Recipes for Megatron-based LLM and VLM models Affected versions of this package are vulnerable to Arbitrary Code Injection via the deserialization process. An attacker can execute arbitrary code, escalate privileges, tamper with data, and...

Use of Externally-Controlled Input to Select Classes or Code ('Unsafe Reflection')

Overview megatron-bridge is a Megatron Bridge: Training Recipes for Megatron-based LLM and VLM models Affected versions of this package are vulnerable to Use of Externally-Controlled Input to Select Classes or Code 'Unsafe Reflection' in the process responsible for dynamically managing code...

Arbitrary Code Injection

Overview megatron-bridge is a Megatron Bridge: Training Recipes for Megatron-based LLM and VLM models Affected versions of this package are vulnerable to Arbitrary Code Injection in the code generation process. An attacker can execute arbitrary code, escalate privileges, tamper with data, and...

CVE-2025-23350

NVIDIA ConnectX and BlueField contain a vulnerability in the command interface where a local user with virtual function VF access may cause a write out of bounds by crafted input. A successful exploit of this vulnerability may lead to arbitrary code execution on the device...

CVE-2025-23351

NVIDIA ConnectX and BlueField contain a vulnerability in the command interface where a local user with virtual function VF access may cause a write out of bounds by crafted input. A successful exploit of this vulnerability may lead to arbitrary code execution on the device...

Adobe Patches 7 CVSS 10.0 Flaws in ColdFusion and Campaign Classic

Adobe has released patches for multiple maximum-severity security flaws impacting Adobe ColdFusion and Adobe Campaign Classic. The ColdFusion updates "resolves critical and important vulnerabilities that could lead to arbitrary code execution, privilege escalation, arbitrary file system read, and...

CVE-2025-23351

NVIDIA ConnectX and BlueField contain a vulnerability in the command interface where a local user with virtual function VF access may cause a write out of bounds by crafted input. A successful exploit of this vulnerability may lead to arbitrary code execution on the device...

CVE-2025-23350

NVIDIA ConnectX and BlueField contain a vulnerability in the command interface where a local user with virtual function VF access may cause a write out of bounds by crafted input. A successful exploit of this vulnerability may lead to arbitrary code execution on the device...

CVE-2025-23350

NVIDIA reports CVE-2025-23350 affecting ConnectX and BlueField through a vulnerability in the command interface that allows a local user with VF access to cause a write out of bounds via crafted input, potentially enabling arbitrary code execution on the device. The issue impacts devices with VF ...

CVE-2025-23350

NVIDIA ConnectX and BlueField contain a vulnerability in the command interface where a local user with virtual function VF access may cause a write out of bounds by crafted input. A successful exploit of this vulnerability may lead to arbitrary code execution on the device...

CVE-2025-23350

NVIDIA ConnectX and BlueField contain a vulnerability in the command interface where a local user with virtual function VF access may cause a write out of bounds by crafted input. A successful exploit of this vulnerability may lead to arbitrary code execution on the device...

Time-of-check Time-of-use (TOCTOU) Race Condition

Overview Affected versions of this package are vulnerable to Time-of-check Time-of-use TOCTOU Race Condition through a race condition in the process. An attacker can execute arbitrary code, escalate privileges, and tamper with data by exploiting the timing window between security checks and use o...

CVE-2026-54672

A flaw was found in electron-updater, a component used for automatic updates in Electron applications. This vulnerability arises because AppImage targets, built by app-builder-lib, incorrectly add the current working directory to the dynamic linker search path when setting the LDLIBRARYPATH...

rrdtool: rrdtool: Stack buffer overflow allows local code execution or denial of service

A flaw was found in rrdcached, a component of rrdtool. A local attacker with access to a rrdcached socket can exploit a stack-based buffer overflow by sending an oversized CREATE request. This vulnerability can lead to a denial of service by crashing the daemon or potentially allow for arbitrary...

Important: Red Hat Security Advisory: Red Hat Ansible Automation Platform 2.6 Product Security and Bug Fix Update

An update is now available for Red Hat Ansible Automation Platform 2.6 Red Hat Product Security has rated this update as having a security impact of Important. A Common Vulnerability Scoring System CVSS base score, which gives a detailed severity rating, is available for each vulnerability from t...

kernel: sctp: revalidate list cursor after sctp_sendmsg_to_asoc() in SCTP_SENDALL

A flaw was found in the Linux kernel's Stream Control Transmission Protocol SCTP implementation. A race condition exists in the SCTPSENDALL path where a cached list entry is not properly revalidated after the socket lock is temporarily released. This allows a local attacker or a remote attacker v...

kernel: dlm: validate length in dlm_search_rsb_tree

A flaw was found in the Linux kernel's Distributed Lock Manager dlm module. An attacker could send specially crafted network messages with an oversized length parameter to the dlmdumprsbname function. This lack of validation can lead to an out-of-bounds write in the dlmsearchrsbtree function,...

postgresql: PostgreSQL: Arbitrary code execution vulnerability in 'refint' module

A flaw was found in PostgreSQL. A stack buffer overflow in the 'refint' module allows an unprivileged database user to execute arbitrary code as the operating system user running the database. Additionally, a separate vulnerability exists where, if an application uses a user-controlled column as ...