Ronen Shustin

An issue exists in Azure Entra ID where external initialization of trusted variables or data stores can allow an unauthorized attacker to elevate privileges locally.

NVIDIA Triton Inference Server for Windows and Linux contains a vulnerability in the Python backend that allows an attacker to cause an out-of-bounds write by sending a specially crafted request. Successful exploitation of this vulnerability may lead to remote code execution, denial of service, data tampering, or information disclosure. The vulnerability chain allows unauthenticated attackers to gain full control of the server.

NVIDIA Triton Inference Server for Windows and Linux contains an issue in the Python backend where an attacker could cause an out-of-bounds read by sending a request. A successful exploit of this issue might lead to information disclosure.

A security issue exists in ingress-nginx where the `mirror-target` and `mirror-host` Ingress annotations can be exploited to inject arbitrary configuration into nginx. Successful exploitation can lead to arbitrary code execution within the context of the ingress-nginx controller and potential disclosure of Secrets accessible to the controller. In a default installation, the controller has access to all Secrets cluster-wide. The issue stems from unsanitized mirror annotations.

A security issue exists where the `auth-tls-match-cn` Ingress annotation can be used to inject configuration into nginx. This may allow a remote attacker to execute arbitrary code within the context of the ingress-nginx controller and disclose Secrets accessible to the controller. In default installations, the controller typically has access to all Secrets cluster-wide.

A security issue exists in the ingress-nginx controller due to insufficient input validation and inadequate spatial separation. An unauthenticated attacker with network access to the pod network can send a specially crafted request to the admission webhook endpoint on port 8443. This allows the attacker to inject malicious NGINX configurations and achieve arbitrary code execution in the context of the controller. Successful exploitation can lead to the disclosure of Kubernetes Secrets accessible to the controller, which, in default installations, may include all Secrets cluster-wide, potentially granting full control over the Kubernetes cluster.

A security issue was discovered in ingress-nginx where attacker-provided data are included in a filename by the ingress-nginx Admission Controller feature, resulting in directory traversal within the container. This could result in denial of service, or when combined with other vulnerabilities, limited disclosure of Secret objects from the cluster. The issue affects over 6,500 clusters and could lead to unauthenticated remote code execution (RCE) or secret theft.

A security issue exists in ingress-nginx where the `auth-url` Ingress annotation can be exploited to inject configuration into nginx. Successful exploitation can lead to arbitrary code execution within the context of the ingress-nginx controller and potential disclosure of Secrets accessible to the controller. In a default installation, the controller has access to all Secrets within the cluster. The issue is due to insufficient input validation.

NVIDIA Container Toolkit and NVIDIA GPU Operator are affected by a Time-of-Check Time-of-Use (TOCTOU) vulnerability. This flaw can allow a crafted container image to gain access to the host file system, potentially leading to code execution, denial of service, escalation of privileges, information disclosure, and data tampering. A bypass was discovered for a previously patched security flaw, reflagged as CVE-2025-23359. The vulnerability exists due to errors in synchronization when using a shared resource, creating a race condition.

The issue is related to an improper isolation vulnerability where a specially crafted container image could lead to modification of a host binary. A successful exploit may lead to code execution, denial of service, escalation of privileges, information disclosure, and data tampering. The "GET /api/v1/status/{job id}" and "POST /api/v1/search" API endpoints are mentioned, with the `job id` variable and `Content-Type` being relevant. No information is provided about the estimated number of potentially affected devices or real-world incidents.

The NVIDIA Container Toolkit contains an improper isolation vulnerability. A specially crafted container image could lead to untrusted code running in the host’s network namespace. This issue is present only when the NVIDIA Container Toolkit is configured in a nondefault way. A successful exploit may lead to denial of service and escalation of privileges.

The NVIDIA Container Toolkit contains an improper isolation vulnerability where a specially crafted container image could lead to untrusted code obtaining read and write access to host devices. This issue is present only when the NVIDIA Container Toolkit is configured in a nondefault way. A successful exploit may lead to code execution, denial of service, escalation of privileges, information disclosure, and data tampering.

The issue is related to errors in processing symbolic links, which can be exploited by a remote attacker to impact data integrity. A specially crafted container image can lead to the creation of unauthorized files on the host, although the name and location of these files cannot be controlled by the attacker. This might result in data tampering.

The issue is related to the default mode of operation in NVIDIA Container Toolkit, allowing a specially crafted container image to create empty files on the host file system. This vulnerability does not impact use cases where CDI is used. A successful exploit may lead to data tampering. The vulnerability is also associated with a null pointer dereference due to concurrent access to a resource, potentially allowing a remote attacker to modify arbitrary data by using a specially crafted container image.

The issue is related to incorrect code generation management in Microsoft Defender for IoT, which can be exploited by a remote attacker to execute arbitrary code using a specially crafted request.

The issue is related to insufficient access control in Microsoft Defender for IoT, which can be exploited to elevate privileges in the system.

The issue is related to incorrect code generation management in Microsoft Defender for IoT, which can be exploited by a remote attacker to execute arbitrary code using a specially crafted request. This may also involve an SQL injection vulnerability in the update-handshake endpoint, potentially allowing authentication bypass.

The issue is related to a lack of protection measures for the SQL query structure in Microsoft Defender for IoT, which can be exploited to execute arbitrary code. This can potentially allow an attacker to bypass authentication.

The issue allows remote attackers to execute arbitrary code or cause a denial of service, resulting in memory corruption and application crash, via specially crafted packets.

The issue is related to a spoofing vulnerability that allows an attacker to bypass authentication. This can be exploited by a remote attacker using a specially crafted request, potentially leading to spoofing attacks. The vulnerability exists due to Azure Stack's failure to properly validate certain requests.