CVE-2025-52496: Mbed TLSan open-source C library design weakness (6th July 2025)

Preface: What is the difference between FreeRTOS and Cmsis-RTOS?

Basically FreeRTOS is a RTOS, while CMSIS-RTOS is only a wrapper for any RTOS (like FreeRTOS, CMSIS-RTOS RTX or anything you want). CMSIS-RTOS is an API that enables consistent software layers with middleware and library components. Mbed TLS aims to provide a set of powerful and flexible cryptographic and security building blocks, mainly for embedded systems, focusing on ease of integration and security. The design objective strives to be lean,  prioritizing readability, documentation and testability, while minimizing dependencies and providing a loosely coupled architecture. This allows developers to integrate only the necessary components without the overhead of the entire library.

Background: Do multi-threaded programs use the same AES key?

Yes, multithreaded programs using AES encryption typically use the same AES key for both encryption and decryption, as AES is a symmetric encryption algorithm. Both the sender and receiver need to share the same secret key to encrypt and decrypt data. In a multithreaded context, each thread would utilize this shared key when performing encryption or decryption operations.

Vulnerability details: Mbed TLS before 3.6.4 has a race condition in AESNI detection if certain compiler optimizations occur. An attacker may be able to extract an AES key from a multithreaded program, or perform a GCM forgery.

Official announcement: Please refer to the link for details – https://nvd.nist.gov/vuln/detail/CVE-2025-52496

CVE-2025-6073 – Industrial Controls Be Aware! (4th July 2025)

Preface: The default configuration of the ABB RMC-100’s REST interface is disabled. ABB recommends leaving the REST interface disabled when not in use, particularly when configuring MQTT functionality. The RMC-100 is not intended for access over public networks.

Background: The ABB RMC-100 is a popular and widely used remote modular controller, particularly within the oil and gas industry. It is known for its scalability and ability to manage automation, liquids and gas measurement, and asset data concentration for various facility sizes, from large production and transmission facilities to smaller systems. The RMC-100 is part of ABB’s Totalflow portfolio, which has seen over 430,000 units sold since the 1980s.

Service available in some Totalflow devices like the RMC-100. When enabled, the device REST server capabilities are enabled. The device then can be accessed by a REST client such as a web browser. The access is for the configuration of the MQTT parameters.

Uses HTTP methods (protocol) to access resources on a REST server. For example, the web browser which accesses the MQTT configuration interface on the RMC-100.

Vulnerability details: Stack-based Buffer Overflow vulnerability in ABB RMC-100, ABB RMC-100 LITE. When the REST interface is enabled by the user, and an attacker gains access to the control network, and user/password broker authentication is enabled, and CVE-2025-6074 is exploited, the attacker can overflow the buffer for username or password.

Affected Products: This issue affects RMC-100: from 2105457-043 through 2105457-045; RMC-100 LITE: from 2106229-015 through 2106229-016.

Official announcement: Please see the link for details –

https://nvd.nist.gov/vuln/detail/CVE-2025-6073

CVE-2025-46647: A vulnerability of plugin openid-connect in Apache APISIX.(3rd July 2025)

Preface: API Gateway can be helpful for ChatGPT plugin developers to expose, secure, manage, and monitor their API endpoints. This repo demonstrates how to use Apache APISIX API Gateway as a front door for communication between ChatGPT custom plugins and backend APIs. For more details, please refer to the link – https://github.com/Boburmirzo/apisix-chatgpt-gateway-plugin

Background: The primary design objective of Apache APISIX is to provide a high-performance, cloud-native API gateway that can handle a large volume of API traffic and microservices, with a focus on flexibility, scalability, and dynamic configuration management. It aims to be a unified proxy infrastructure for various scenarios like API management, service mesh, and ingress control.

The OpenID Connect (OIDC) plugin for Apache APISIX enables centralized authentication for APIs by integrating with OpenID Connect providers (like Okta, Auth0, Keycloak). It allows users to authenticate through a designated provider and then access APIs through APISIX. The plugin handles the redirection to the provider’s login page, token exchange, and passing user information to the upstream services.

Vulnerability details: A vulnerability of plugin openid-connect in Apache APISIX. This vulnerability will only have an impact if all of the following conditions are met:

1. Use the openid-connect plugin with introspection mode

2. The auth service connected to openid-connect provides services to multiple issuers

3. Multiple issuers share the same private key and relies only on the issuer being different If affected by this vulnerability, it would allow an attacker with a valid account on one of the issuers to log into the other issuer.

Remedy: This issue affects Apache APISIX: until 3.12.0. Users are recommended to upgrade to version 3.12.0 or higher.

Official announcement: For more details, please refer to the link –

https://nvd.nist.gov/vuln/detail/CVE-2025-46647

CVE-2025-0038 exposes a runtime vulnerability due to missing checks in PMU firmware. (2nd July 2025)

Preface: Users typically build custom PMU firmware tailored to their specific hardware platform and application requirements.

PMU firmware can be loaded by either FSBL or CSU BootROM (CBR). Both these flows are supported by AMD. Loading PMU firmware using FSBL has the following benefits:

– Possible quick boot time, when PMU firmware is loaded after bitstream.

– In use cases where you want two BIN files – stable and upgradable, PMU firmware can be part of the upgradable (by FSBL) image.

Background: The primary design objective of AMD’s Zynq™ UltraScale+™ devices is to provide a highly integrated platform that combines the processing power of a multi-core ARM processor with the flexibility of programmable logic (FPGA fabric). This enables a wide range of applications by offering both real-time control and processing capabilities within a single chip. The devices also prioritize low power consumption, security features, and efficient memory management.

Ref: Arm Trusted Firmware (ATF) and its role in managing the Secure Monitor and Trusted Board Boot Requirements (TBBR). These are essential for establishing a secure boot process and managing transitions between the secure and non-secure worlds in Arm-based systems like the Zynq UltraScale+.

Vulnerability details: In AMD Zynq UltraScale+ devices, the lack of address validation when executing CSU runtime services through the PMU Firmware can allow access to isolated or protected memory spaces resulting in the loss of integrity and confidentiality.

Official announcement: Please see the link for details –

https://www.amd.com/en/resources/product-security/bulletin/amd-sb-8008.html

https://docs.amd.com/r/en-US/000037628/Affected-Products

CVE-2025-49521: Red Hat Ansible Automation Platform 2.5 Product Security and Bug Fix Update (1st July 2025)

Preface: Ansible Automation Platform is a broader enterprise automation platform designed to manage and automate various IT operations, including infrastructure, cloud, networking, and security. While it can be used for automating web server deployments and configurations. Besides, web hosting service providers can and often do use the Ansible Automation Platform for automating various tasks related to web hosting and infrastructure management.

Background: In Ansible, Jinja2 templating is widely used to dynamically render variables, expressions, and logic in playbooks, templates, and even hooks (like webhooks or event triggers in EDA).

You can use Jinja2 in:

•       Playbooks: For dynamic task names, conditions, and variables.

•       Templates: To generate configuration files.

•       Hooks or Webhooks: Especially in EDA, where incoming payloads can be parsed and matched using Jinja2 expressions.

When a POST request is sent to http[:][//]<EDA_HOST>[:]5000/alert with the payload:

•       The EDA controller receives the event.

•       It evaluates the condition using Jinja2.

•       If matched, it runs the playbook respond_to_critical_alert[.]yml.

Vulnerability details: A flaw was found in the EDA component of the Ansible Automation Platform, where user-supplied Git branch or refspec values are evaluated as Jinja2 templates. This vulnerability allows authenticated users to inject expressions that execute commands or access sensitive files on the EDA worker. In OpenShift, it can lead to service account token theft.

Official announcement: Please refer to the link for details – https://nvd.nist.gov/vuln/detail/CVE-2025-49521