Vulnerability | Breaking Cybersecurity News | The Hacker News

Cybersecurity researchers have disclosed two security flaws in Google's Vertex machine learning (ML) platform that, if successfully exploited, could allow malicious actors to escalate privileges and exfiltrate models from the cloud. "By exploiting custom job permissions, we were able to escalate our privileges and gain unauthorized access to all data services in the project," Palo Alto Networks Unit 42 researchers Ofir Balassiano and Ofir Shaty said in an analysis published earlier this week. "Deploying a poisoned model in Vertex AI led to the exfiltration of all other fine-tuned models, posing a serious proprietary and sensitive data exfiltration attack risk." Vertex AI is Google's ML platform for training and deploying custom ML models and artificial intelligence (AI) applications at scale. It was first introduced in May 2021. Crucial to leveraging the privilege escalation flaw is a feature called Vertex AI Pipelines , which allows users to automat...

Cybersecurity researchers have disclosed a high-severity security flaw in the PostgreSQL open-source database system that could allow unprivileged users to alter environment variables, and potentially lead to code execution or information disclosure. The vulnerability, tracked as CVE-2024-10979 , carries a CVSS score of 8.8. Environment variables are user-defined values that can allow a program to dynamically fetch various kinds of information, such as access keys and software installation paths, during runtime without having to hard-code them. In certain operating systems, they are initialized during the startup phase. "Incorrect control of environment variables in PostgreSQL PL/Perl allows an unprivileged database user to change sensitive process environment variables (e.g., PATH )," PostgreSQL said in an advisory released Thursday. "That often suffices to enable arbitrary code execution, even if the attacker lacks a database server operating system user." ...

IT leaders know the drill—regulators and cyber insurers demand regular network penetration testing to keep the bad guys out. But here's the thing: hackers don't wait around for compliance schedules. Most companies approach network penetration testing on a set schedule, with the most common frequency being twice a year (29%), followed by three to four times per year (23%) and once per year (20%), according to the Kaseya Cybersecurity Survey Report 2024 . Compliance-focused testing can catch vulnerabilities that exist at the exact time of testing, but it's not enough to stay ahead of attackers in a meaningful way. Why More Frequent Testing Makes Sense When companies test more often, they're not just checking a box for compliance—they're actually protecting their networks. The Kaseya survey also points out that the top drivers for network penetration testing are: Cybersecurity Control and Validation (34%) – ensuring the security controls work and vulnerabilities are minimized. Re...

The U.S. Cybersecurity and Infrastructure Security Agency (CISA) on Thursday warned that two more flaws impacting the Palo Alto Networks Expedition software have come under active exploitation in the wild. To that end, it has added the vulnerabilities to its Known Exploited Vulnerabilities ( KEV ) catalog, requiring Federal Civilian Executive Branch (FCEB) agencies to apply the necessary updates by December 5, 2024. The security flaws are listed below - CVE-2024-9463 (CVSS score: 9.9) - Palo Alto Networks Expedition OS Command Injection Vulnerability CVE-2024-9465 (CVSS score: 9.3) - Palo Alto Networks Expedition SQL Injection Vulnerability Successful exploitation of the vulnerabilities could allow an unauthenticated attacker to run arbitrary OS commands as root in the Expedition migration tool or reveal its database contents. This could then pave the way for disclosure of usernames, cleartext passwords, device configurations, and device API keys of PAN-OS firewalls, or cr...

A newly patched security flaw impacting Windows NT LAN Manager (NTLM) was exploited as a zero-day by a suspected Russia-linked actor as part of cyber attacks targeting Ukraine. The vulnerability in question, CVE-2024-43451 (CVSS score: 6.5), refers to an NTLM hash disclosure spoofing vulnerability that could be exploited to steal a user's NTLMv2 hash. It was patched by Microsoft earlier this week. "Minimal interaction with a malicious file by a user such as selecting (single-click), inspecting (right-click), or performing an action other than opening or executing could trigger this vulnerability," Microsoft revealed in its advisory. Israeli cybersecurity company ClearSky, which discovered the zero-day exploitation of the flaw in June 2024, said it's been abused as part of an attack chain that delivers the open-source Spark RAT malware. "The vulnerability activates URL files, leading to malicious activity," the company said, adding the malicious file...

A security analysis of the OvrC cloud platform has uncovered 10 vulnerabilities that could be chained to allow potential attackers to execute code remotely on connected devices. "Attackers successfully exploiting these vulnerabilities can access, control, and disrupt devices supported by OvrC; some of those include smart electrical power supplies, cameras, routers, home automation systems, and more," Claroty researcher Uri Katz said in a technical report. Snap One's OvrC, pronounced "oversee," is advertised as a "revolutionary support platform" that enables homeowners and businesses to remotely manage, configure, and troubleshoot IoT devices on the network. According to its website, OvrC solutions are deployed at over 500,000 end-user locations. According to a coordinated advisory issued by the U.S. Cybersecurity and Infrastructure Security Agency (CISA), successful exploitation of the identified vulnerabilities could allow an attacker to ...

Microsoft on Tuesday revealed that two security flaws impacting Windows NT LAN Manager ( NTLM ) and Task Scheduler have come under active exploitation in the wild. The security vulnerabilities are among the 90 security bugs the tech giant addressed as part of its Patch Tuesday update for November 2024. Of the 90 flaws, four are rated Critical, 85 are rated Important, and one is rated Moderate in severity. Fifty-two of the patched vulnerabilities are remote code execution flaws. The fixes are in addition to 31 vulnerabilities Microsoft resolved in its Chromium-based Edge browser since the release of the October 2024 Patch Tuesday update. The two vulnerabilities that have been listed as actively exploited are below - CVE-2024-43451 (CVSS score: 6.5) - Windows NTLM Hash Disclosure Spoofing Vulnerability CVE-2024-49039 (CVSS score: 8.8) - Windows Task Scheduler Elevation of Privilege Vulnerability "This vulnerability discloses a user's NTLMv2 hash to the attacker who c...

Cybersecurity researchers have disclosed new security flaws impacting Citrix Virtual Apps and Desktop that could be exploited to achieve unauthenticated remote code execution (RCE) The issue, per findings from watchTowr , is rooted in the Session Recording component that allows system administrators to capture user activity, and record keyboard and mouse input, along with a video stream of the desktop for audit, compliance, and troubleshooting purposes. Particularly, the vulnerability exploits the "combination of a carelessly-exposed MSMQ instance with misconfigured permissions that leverages BinaryFormatter can be reached from any host via HTTP to perform unauthenticated RCE," security researcher Sina Kheirkhah said. The vulnerability details are listed below - CVE-2024-8068 (CVSS score: 5.1) - Privilege escalation to NetworkService Account access CVE-2024-8069 (CVSS score: 5.1) - Limited remote code execution with the privilege of a NetworkService Account acces...

Cybersecurity researchers have uncovered nearly two dozen security flaws spanning 15 different machine learning (ML) related open-source projects. These comprise vulnerabilities discovered both on the server- and client-side, software supply chain security firm JFrog said in an analysis published last week. The server-side weaknesses "allow attackers to hijack important servers in the organization such as ML model registries, ML databases and ML pipelines," it said . The vulnerabilities, discovered in Weave, ZenML, Deep Lake, Vanna.AI, and Mage AI, have been broken down into broader sub-categories that allow for remotely hijacking model registries, ML database frameworks, and taking over ML Pipelines. A brief description of the identified flaws is below - CVE-2024-7340 (CVSS score: 8.8) - A directory traversal vulnerability in the Weave ML toolkit that allows for reading files across the whole filesystem, effectively allowing a low-privileged authenticated user to es...

Hewlett Packard Enterprise (HPE) has released security updates to address multiple vulnerabilities impacting Aruba Networking Access Point products, including two critical bugs that could result in unauthenticated command execution. The flaws affect Access Points running Instant AOS-8 and AOS-10 - AOS-10.4.x.x: 10.4.1.4 and below Instant AOS-8.12.x.x: 8.12.0.2 and below Instant AOS-8.10.x.x: 8.10.0.13 and below The most severe among the six newly patched vulnerabilities are CVE-2024-42509 (CVSS score: 9.8) and CVE-2024-47460 (CVSS score: 9.0), two critical unauthenticated command injection flaws in the CLI Service that could result in the execution of arbitrary code. "Command injection vulnerability in the underlying CLI service could lead to unauthenticated remote code execution by sending specially crafted packets destined to the PAPI (Aruba's Access Point management protocol) UDP port (8211)," HPE said in an advisory for both the flaws. "Successful exp...

Cybersecurity researchers have discovered a new phishing campaign that spreads a new fileless variant of known commercial malware called Remcos RAT . Remcos RAT "provides purchases with a wide range of advanced features to remotely control computers belonging to the buyer," Fortinet FortiGuard Labs researcher Xiaopeng Zhang said in an analysis published last week. "However, threat actors have abused Remcos to collect sensitive information from victims and remotely control their computers to perform further malicious acts." The starting point of the attack is a phishing email that uses purchase order-themed lures to convince recipients to open a Microsoft Excel attachment. The malicious Excel document is designed to exploit a known remote code execution flaw in Office ( CVE-2017-0199 , CVSS score: 7.8) to download an HTML Application (HTA) file ("cookienetbookinetcahce.hta") from a remote server ("192.3.220[.]22") and launch it using mshta.e...

Palo Alto Networks on Friday issued an informational advisory urging customers to ensure that access to the PAN-OS management interface is secured because of a potential remote code execution vulnerability. "Palo Alto Networks is aware of a claim of a remote code execution vulnerability via the PAN-OS management interface," the company said . "At this time, we do not know the specifics of the claimed vulnerability. We are actively monitoring for signs of any exploitation." In the interim, the network security vendor has recommended that users correctly configure the management interface in line with the best practices, and make sure that access to it is possible only via trusted internal IPs to limit the attack surface. It goes without saying that the management interface should not be exposed to the Internet. Some of the other guidelines to reduce exposure are listed below - Isolate the management interface on a dedicated management VLAN Use jump servers...

The threat actors behind the AndroxGh0st malware are now exploiting a broader set of security flaws impacting various internet-facing applications, while also deploying the Mozi botnet malware. "This botnet utilizes remote code execution and credential-stealing methods to maintain persistent access, leveraging unpatched vulnerabilities to infiltrate critical infrastructures," CloudSEK said in a new report. AndroxGh0st is the name given to a Python-based cloud attack tool that's known for its targeting of Laravel applications with the goal of sensitive data pertaining to services like Amazon Web Services (AWS), SendGrid, and Twilio. Active since at least 2022, it has previously leveraged flaws in the Apache web server ( CVE-2021-41773 ), Laravel Framework ( CVE-2018-15133 ), and PHPUnit ( CVE-2017-9841 ) to gain initial access, escalate privileges, and establish persistent control over compromised systems. Earlier this January, U.S. cybersecurity and intelligence a...

The U.S. Cybersecurity and Infrastructure Security Agency (CISA) on Thursday added a now-patched critical security flaw impacting Palo Alto Networks Expedition to its Known Exploited Vulnerabilities ( KEV ) catalog, citing evidence of active exploitation. The vulnerability, tracked as CVE-2024-5910 (CVSS score: 9.3), concerns a case of missing authentication in the Expedition migration tool that could lead to an admin account takeover. "Palo Alto Expedition contains a missing authentication vulnerability that allows an attacker with network access to takeover an Expedition admin account and potentially access configuration secrets, credentials, and other data," CISA said in an alert. The shortcoming impacts all versions of Expedition prior to version 1.2.92, which was released in July 2024 to plug the problem. There are currently no reports on how the vulnerability is being weaponized in real-world attacks, but Palo Alto Networks has since revised its original adviso...

Cisco has released security updates to address a maximum severity security flaw impacting Ultra-Reliable Wireless Backhaul ( URWB ) Access Points that could permit unauthenticated, remote attackers to run commands with elevated privileges. Tracked as CVE-2024-20418 (CVS score: 10.0), the vulnerability has been described as stemming from a lack of input validation to the web-based management interface of the Cisco Unified Industrial Wireless Software. "An attacker could exploit this vulnerability by sending crafted HTTP requests to the web-based management interface of an affected system," Cisco said in an advisory released Wednesday. "A successful exploit could allow the attacker to execute arbitrary commands with root privileges on the underlying operating system of the affected device." The shortcoming impacts following Cisco products in scenarios where the URWB operating mode is enabled - Catalyst IW9165D Heavy Duty Access Points Catalyst IW9165E Rugge...

Cybersecurity researchers have discovered a malicious package on the Python Package Index (PyPI) that has racked up thousands of downloads for over three years while stealthily exfiltrating developers' Amazon Web Services (AWS) credentials. The package in question is " fabrice ," which typosquats a popular Python library known as " fabric ," which is designed to execute shell commands remotely over SSH.  While the legitimate package has over 202 million downloads, its malicious counterpart has been downloaded more than 37,100 times to date. As of writing, "fabrice" is still available for download from PyPI. It was first published in March 2021. The typosquatting package is designed to exploit the trust associated with "fabric," incorporating "payloads that steal credentials, create backdoors, and execute platform-specific scripts," security firm Socket said . "Fabrice" is designed to carry out its malicious actions ...

Taiwanese network-attached storage (NAS) appliance maker Synology has addressed a critical security flaw impacting DiskStation and BeePhotos that could lead to remote code execution. Tracked as CVE-2024-10443 and dubbed RISK:STATION by Midnight Blue, the zero-day flaw was demonstrated at the Pwn2Own Ireland 2024 hacking contest by security researcher Rick de Jager. RISK:STATION is an "unauthenticated zero-click vulnerability allowing attackers to obtain root-level code execution on the popular Synology DiskStation and BeeStation NAS devices, affecting millions of devices," the Dutch company said . The zero-click nature of the vulnerability means it does not require any user interaction to trigger the exploitation, thereby allowing attackers to gain access to the devices to steal sensitive data and plant additional malware. The flaw impacts the following versions - BeePhotos for BeeStation OS 1.0 (Upgrade to 1.0.2-10026 or above) BeePhotos for BeeStation OS 1.1 (Upgr...

Google has warned that a security flaw impacting its Android operating system has come under active exploitation in the wild. The vulnerability, tracked as CVE-2024-43093, has been described as a privilege escalation flaw in the Android Framework component that could result in unauthorized access to "Android/data," "Android/obb," and "Android/sandbox" directories, and their respective sub-directories, according to a code commit message . There are currently no details about how the vulnerability is being weaponized in real-world attacks, but Google acknowledged in its monthly bulletin that there are indications it "may be under limited, targeted exploitation." The tech giant has also flagged CVE-2024-43047, a now-patched security bug in Qualcomm chipsets, as having been actively exploited. A use-after-free vulnerability in the Digital Signal Processor (DSP) Service, a successful exploitation of the security flaw could lead to memory corrupti...