#vulnerability

### Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-mh55-gqvf-xfwm. This link is maintained to preserve external references. ### Original Description Middleware causes a prohibitive amount of heap allocations when processing malicious preflight requests that include a Access-Control-Request-Headers (ACRH) header whose value contains many commas. This behavior can be abused by attackers to produce undue load on the middleware/server as an attempt to cause a denial of service.

The now-patched vulnerabilities exist at the firmware level and enable deep persistence on compromised systems.

A vulnerability was found in Keycloak-services. Special characters used during e-mail registration may perform SMTP Injection and unexpectedly send short unwanted e-mails. The email is limited to 64 characters (limited local part of the email), so the attack is limited to very shorts emails (subject and little data, the example is 60 chars). This flaw's only direct consequence is an unsolicited email being sent from the Keycloak server. However, this action could be a precursor for more sophisticated attacks.

Two critical vulnerabilities affect the security vendor's management console, one of which is under active exploitation. The company has updated cloud-based products but won't have a patch for its on-premises version until mid-August.

For likely the first time ever, security researchers have shown how AI can be hacked to create real world havoc, allowing them to turn off lights, open smart shutters, and more.

A Regular Expression Denial of Service (ReDoS) vulnerability exists in the Hugging Face Transformers library, specifically in the `convert_tf_weight_name_to_pt_weight_name()` function. This function, responsible for converting TensorFlow weight names to PyTorch format, uses a regex pattern `/[^/]*___([^/]*)/` that can be exploited to cause excessive CPU consumption through crafted input strings due to catastrophic backtracking. The vulnerability affects versions up to 4.51.3 and is fixed in version 4.53.0. This issue can lead to service disruption, resource exhaustion, and potential API service vulnerabilities, impacting model conversion processes between TensorFlow and PyTorch formats.

Vault and Vault Enterprise’s (“Vault”) ldap auth method may not have correctly enforced MFA if username_as_alias was set to true and a user had multiple CNs that are equal but with leading or trailing spaces. Fixed in Vault Community Edition 1.20.2 and Vault Enterprise 1.20.2, 1.19.8, 1.18.13, and 1.16.24.

GuidePoint Security uncovers a new Akira ransomware tactic targeting SonicWall VPNs. The group's use of drivers to disable defenses is a significant threat to businesses.

Cisco Talos’ Vulnerability Discovery & Research team recently disclosed seven vulnerabilities in WWBN AVideo, four in MedDream, and one in an Eclipse ThreadX module. The vulnerabilities mentioned in this blog post have been patched by their respective vendors, all in adherence to Cisco’s third-party vulnerability disclosure policy

Disclosure: This article was provided by ANY.RUN. The information and analysis presented are based on their research and findings.