01/31/2024-02/07/2024 Critical JetBrains Security Flaw, New Flaws in Azure, Cloudflare Breach And More.

1. Critical JetBrains TeamCity On-Premises Flaw Exposes Servers to Takeover – Patch Now

JetBrains warns of a critical security flaw (CVE-2024-23917) in its TeamCity On-Premises software, scoring 9.8 out of 10 on the CVSS scale. This flaw could allow threat actors to seize control of vulnerable instances. Attackers with HTTP(S) access might bypass authentication and gain administrative control. The vulnerability affects versions from 2017.1 to 2023.11.2, fixed in 2023.11.3. Discovered by an external researcher on January 19, 2024, users can apply a security patch plugin if unable to update. JetBrains suggests temporarily blocking public access for servers unable to update immediately. While there’s no known exploitation, a similar flaw (CVE-2023-42793) saw active exploitation by ransomware gangs and state-sponsored groups after disclosure last year.

2. Experts Detail New Flaws in Azure HDInsight Spark, Kafka, and Hadoop Services

Three security vulnerabilities in Azure HDInsight’s Apache Hadoop, Kafka, and Spark services allow privilege escalation and a regular expression denial-of-service (ReDoS).

Orca security researcher Lidor Ben Shitrit reported the flaws, including:

• CVE-2023-36419 (CVSS: 8.8) – Apache Oozie XXE Injection Elevation of Privilege;
• CVE-2023-38156 (CVSS: 7.2) – Apache Ambari JDBC Injection Elevation of Privilege;
• Apache Oozie ReDoS Vulnerability.

Attackers could exploit privilege escalation flaws to gain administrator privileges by crafting network requests. The XXE flaw permits root-level file reading and privilege escalation, while the JDBC injection flaw facilitates obtaining a reverse shell as root. The ReDoS vulnerability arises from inadequate input validation, enabling attackers to trigger a denial-of-service by causing an intensive loop operation. Microsoft has released fixes on October 26, 2023, following responsible disclosure. Exploitation could disrupt system operations, degrade performance, and impact service availability and reliability.

3. Recent SSRF Flaw in Ivanti VPN Products Undergoes Mass Exploitation

The Ivanti Connect Secure and Policy Secure products face mass exploitation of a disclosed SSRF vulnerability (CVE-2024-21893, CVSS: 8.2), allowing unauthorized access to restricted resources. Shadowserver Foundation noted over 170 IP addresses attempting to establish a reverse shell. The flaw, also affecting Neurons for ZTA, enables attackers to bypass authentication. Exploitation has surged since a proof-of-concept (PoC) exploit by Rapid7. Combining it with a previously patched command injection flaw (CVE-2024-21887), unauthenticated remote code execution is achieved. Notably, CVE-2024-21893 (alias CVE-2023-36661) is in the Shibboleth XMLTooling library, fixed in June 2023. Additionally, Ivanti VPN appliances use outdated open-source components, exposing vulnerabilities. Ivanti has released a second mitigation and begun patching officially. Threat actors exploit CVE-2023-46805 and CVE-2024-21887 for deploying custom web shells named BUSHWALK, CHAINLINE, FRAMESTING, and LIGHTWIRE, as revealed by Mandiant.

4. Cloudflare Breach: Nation-State Hackers Access Source Code and Internal Docs

Cloudflare discloses a likely nation-state attack, revealing unauthorized access to its Atlassian server, leading to document and source code exposure between November 14 and 24, 2023. The sophisticated actor aimed to persistently infiltrate Cloudflare’s network. In response, Cloudflare rotated over 5,000 production credentials, segmented systems, and conducted forensic triage on 4,893 systems. During the four-day reconnaissance period, the attacker accessed Atlassian Confluence and Jira portals, then established persistent access, ultimately accessing 120 code repositories, of which 76 were exfiltrated. Repositories pertained to backups, network configuration, identity management, remote access, and infrastructure management tools. The attacker unsuccessfully tried to breach a console server in São Paulo. Stolen credentials from Okta’s support system, including AWS and Atlassian, facilitated the intrusion. Cloudflare failed to rotate these credentials promptly but terminated malicious connections on November 24, 2023, and engaged CrowdStrike for an independent assessment.

5. FritzFrog Returns with Log4Shell and PwnKit, Spreading Malware Inside Your Network

The FritzFrog threat actor has reemerged with a new variant utilizing the Log4Shell vulnerability to spread within compromised networks. Akamai reports brute-force exploitation targeting vulnerable Java applications. Initially focusing on weak SSH credentials, FritzFrog has expanded its targets to healthcare, education, and government sectors, deploying cryptocurrency miners. Unlike prior versions, the latest version targets internal hosts, leveraging unpatched systems. This shift exploits neglected internal machines, increasing infection risks. FritzFrog enhances its SSH brute-force tactic and utilizes CVE-2021-4034 for local privilege escalation. To evade detection, it avoids dropping files to disk, utilizing /dev/shm and memfd_create for memory-resident payloads. This strategy mirrors techniques used by other Linux-based malware.

6. Exposed Docker APIs Under Attack in ‘Commando Cat’ Cryptojacking Campaign

Docker API endpoints face a cryptojacking campaign named Commando Cat, deploying benign containers via the Commando project. Active since 2024, it’s the second such campaign within months. Targeting Docker hosts, it deploys XMRig and 9Hits Viewer. Commando Cat breaches Docker instances to execute payloads, including backdoors and miners. It checks for specific active services before proceeding. Payloads are delivered from the C2 server, adding SSH keys, creating rogue users, and exfiltrating credentials. Using curl or wget, payloads are fetched and executed, with /dev/shm used for evasion. Forensics are complicated as artifacts avoid disk touch. The attack concludes with a Base64-encoded script deploying XMRig after eliminating competing miners.

01/24/2024-01/31/2024 Upgrade GitLab, Urgent Junos OS Updates, Critical Jenkins Vulnerability, Malicious PyPI Packages And More.

1. URGENT: Upgrade GitLab – Critical Workspace Creation Flaw Allows File Overwrite

GitLab once again released fixes to address a critical security flaw in its Community Edition (CE) and Enterprise Edition (EE) that could be exploited to write arbitrary files while creating a workspace. Tracked as CVE-2024-0402, the vulnerability has a CVSS score of 9.9 out of a maximum of 10. The company also noted patches for the bug have been backported to 16.5.8, 16.6.6, 16.7.4, and 16.8.1. Also resolved by GitLab are four medium-severity flaws that could lead to a regular expression denial-of-service (ReDoS), HTML injection, and the disclosure of a user’s public email address via the tags RSS feed. The latest update arrives two weeks after the DevSecOps platform shipped fixes to close out two critical shortcomings, including one that could be exploited to take over accounts without requiring any user interaction (CVE-2023-7028, CVSS score: 10.0). Users are advised to upgrade the installations to a patched version as soon as possible to mitigate potential risks.

2. Juniper Networks Releases Urgent Junos OS Updates for High-Severity Flaws

Juniper Networks has issued critical updates for SRX Series and EX Series, targeting high-severity vulnerabilities in J-Web (CVE-2024-21619 and CVE-2024-21620). These flaws could empower threat actors to seize control of vulnerable systems. CVE-2024-21619 poses a moderate risk (CVSS score: 5.3) due to a missing authentication vulnerability, exposing sensitive configuration information. On the other hand, CVE-2024-21620 presents a higher risk (CVSS score: 8.8) as a cross-site scripting (XSS) vulnerability, enabling the execution of arbitrary commands. As a temporary measure, Juniper advises users to disable J-Web or limit access to trusted hosts until the updates are implemented. Additionally, two earlier disclosed vulnerabilities (CVE-2023-36846 and CVE-2023-36851) were previously flagged as actively exploited.

3. Critical Jenkins Vulnerability Exposes Servers to RCE Attacks – Patch ASAP!

Jenkins, the open-source CI/CD automation software, has patched nine security flaws, including CVE-2024-23897, a critical bug allowing remote code execution through an arbitrary file read vulnerability in the CLI. Jenkins uses the args4j library for CLI command processing, enabling a feature (expandAtFiles) that replaces ‘@’ followed by a file path with the file’s content. This feature, active by default in Jenkins 2.441 and earlier, could be exploited by threat actors with “Overall/Read” permission to read entire files, potentially leading to various attacks, including remote code execution, XSS, and CSRF protection bypass. SonarSource researcher Yaniv Nizry discovered the flaw, fixed in Jenkins 2.442 and LTS 2.426.3 by disabling the command parser feature. While awaiting the patch, users are advised to disable CLI access as a temporary measure. Proof-of-concept exploits for CVE-2024-23897 have been published, emphasizing the urgency of updates.

4. Malicious PyPI Packages Slip WhiteSnake InfoStealer Malware onto Windows Machines

Researchers have detected malicious packages on the Python Package Index (PyPI), distributing WhiteSnake Stealer malware on Windows systems. These packages, uploaded by a threat actor named “WS,” include nigpal, figflix, telerer, seGMM, fbdebug, sGMM, myGens, NewGends, and TestLibs111. The packages embed Base64-encoded source code in their setup.py files, delivering the final payload upon installation based on the victim’s operating system. WhiteSnake, primarily targeting Windows, has an Anti-VM mechanism, communicates via Tor, and steals information from browsers, cryptocurrency wallets, and various applications. PYTA31, the threat actor tracked by Checkmarx, aims to exfiltrate sensitive and crypto wallet data. Some packages incorporate clipper functionality to replace clipboard content for unauthorized transactions. This discovery highlights the ability of a single malware author to disseminate multiple info-stealing packages into PyPI with distinct payload intricacies.

5. Researchers Uncover How Outlook Vulnerability Could Leak Your NTLM Passwords

A recently patched security flaw in Microsoft Outlook (CVE-2023-35636, CVSS score: 6.5) exposed NT LAN Manager (NTLM) v2 hashed passwords. This vulnerability, addressed in Microsoft’s December 2023 Patch Tuesday updates, allowed threat actors to access passwords when victims opened a specially crafted file. In email attacks, the attacker sends the file, while in web-based attacks, a malicious website hosts it. The flaw originates from the calendar-sharing function in Outlook, utilizing crafted headers. Varonis researcher Dolev Taler discovered the bug, highlighting the potential leakage of NTLM hashes via Windows Performance Analyzer (WPA) and Windows File Explorer, yet these methods remain unpatched. This disclosure coincides with Check Point’s revelation of “forced authentication,” demonstrating the exploitation of NTLM tokens by tricking users into opening a rogue Microsoft Access file.

6. Critical Cisco Flaw Lets Hackers Remotely Take Over Unified Comms Systems

Cisco has issued patches for a critical vulnerability (CVE-2024-20253, CVSS score: 9.9) affecting Unified Communications and Contact Center Solutions. The flaw arises from improper processing of user-provided data, enabling a remote attacker to execute arbitrary code on the target device. Successful exploitation could lead to arbitrary command execution with web services user privileges and potential root access. The affected products include Unified Communications Manager, Unified Communications Manager IM & Presence Service, Unified Communications Manager Session Management Edition, Unified Contact Center Express, Unity Connection, and Virtualized Voice Browser. Synacktiv researcher Julien Egloff discovered the vulnerability. While no workarounds exist, Cisco recommends implementing access control lists (ACLs) as a temporary measure.

01/17/2024-01/24/2024 Malicious NPM Packages, ~40,000 Attacks in 3 Days, MavenGate Attack Could Let Hackers Hijack Java And More.

1. Patch Your GoAnywhere MFT Immediately – Critical Flaw Lets Anyone Be Admin

A critical security flaw has been disclosed in Fortra’s GoAnywhere Managed File Transfer (MFT) software that could be abused to create a new administrator user. Tracked as CVE-2024-0204, the issue carries a CVSS score of 9.8 out of 10. Authentication bypass in Fortra’s GoAnywhere MFT prior to 7.4.1 allows an unauthorized user to create an admin user via the administration portal. The issue is the result of a path traversal weakness in the “/InitialAccountSetup.xhtml” endpoint that could be exploited to create administrative users. Users who cannot upgrade to version 7.4.1 can apply temporary workarounds in non-container deployments by deleting the InitialAccountSetup.xhtml file in the install directory and restarting the services.

2. Malicious NPM Packages Exfiltrate Hundreds of Developer SSH Keys via GitHub

Malicious npm packages, warbeast2000 and kodiak2k, discovered using GitHub to store stolen Base64-encrypted SSH keys from developers. Uploaded at the beginning of the month, the packages attracted 412 and 1,281 downloads before npm took them down on January 21, 2024. Security firm ReversingLabs revealed eight versions of warbeast2000 and over 30 versions of kodiak2k. Both execute postinstall scripts, with warbeast2000 attempting to access private SSH keys and kodiak2k searching for a key named “meow.” Lucija Valentić, a security researcher, explained that warbeast2000 uploads the key to an attacker-controlled GitHub repo. Kodiak2k’s later versions execute a script from an archived GitHub project, launching the Mimikatz tool to extract credentials. Valentić emphasizes this incident showcases cybercriminals exploiting open-source package managers for malicious software supply chain attacks on development and end-user organizations.

3. ~40,000 Attacks in 3 Days: Critical Confluence RCE Under Active Exploitation

Malicious actors have begun to actively exploit a recently disclosed critical security flaw impacting Atlassian Confluence Data Center and Confluence Server. Tracked as CVE-2023-22527 (CVSS score: 10.0), the vulnerability impacts out-of-date versions of the software, allowing unauthenticated attackers to achieve remote code execution on susceptible installations. The shortcoming affects Confluence Data Center and Server 8 versions released before December 5, 2023, as well as 8.4.5. Over 11,000 Atlassian instances have been found to be accessible over the internet as of January 21, 2024, although it’s currently not known how many of them are vulnerable to CVE-2023-22527. This vulnerability has the potential to permit unauthenticated attackers to inject OGNL expressions into the Confluence instance, thereby enabling the execution of arbitrary code and system commands. 

4. PixieFail UEFI Flaws Expose Millions of Computers to RCE, DoS, and Data Theft

Multiple security vulnerabilities, collectively named PixieFail by Quarkslab, have been revealed in the TCP/IP network protocol stack of the widely used Unified Extensible Firmware Interface (UEFI) specification. These nine issues, found in the TianoCore EFI Development Kit II (EDK II), can lead to remote code execution, denial-of-service (DoS), DNS cache poisoning, and sensitive information leakage. UEFI firmware from major providers like AMI, Intel, Insyde, and Phoenix Technologies are affected. The vulnerabilities stem from overflow bugs, out-of-bounds reads, infinite loops, and a weak pseudorandom number generator in EDK II’s NetworkPkg, impacting PXE functionality during the Preboot eXecution Environment (PXE) stage. The vulnerabilities have specific identifiers (CVEs) and varying CVSS scores. The impact and exploitability depend on firmware builds and default PXE boot configurations, with potential for remote code execution, DoS, DNS cache poisoning, or data extraction.

5. MavenGate Attack Could Let Hackers Hijack Java and Android via Abandoned Libraries

Numerous widely-used Java and Android libraries are vulnerable to a novel software supply chain attack called MavenGate. Oversecured reported that domain name purchases could be used to hijack project access, exploiting default build configurations and making attacks difficult to detect. The attack can inject malicious code into dependencies, compromising the build process via a malicious plugin. All Maven-based technologies, including Gradle, are susceptible, affecting over 200 companies like Google, Facebook, and Amazon. Apache Maven, vital for Java projects, is a target due to potentially compromised dependencies in public repositories. The attack involves obtaining expired domains, asserting rights through a DNS TXT record, and gaining access to vulnerable groupIds. While Maven Central believes the outlined attack is infeasible due to automation, Oversecured suggests developers and end-users play crucial roles in ensuring security for direct and transitive dependencies, emphasizing the responsibility of both parties.

6. U.S. Cybersecurity Agency Warns of Actively Exploited Ivanti EPMM Vulnerability

The U.S. Cybersecurity and Infrastructure Security Agency (CISA) has added a critical flaw, CVE-2023-35082, affecting Ivanti Endpoint Manager Mobile (EPMM) and MobileIron Core to its Known Exploited Vulnerabilities catalog. The flaw, now patched, is an authentication bypass that could potentially expose users’ personal information. Ivanti disclosed the vulnerability in August 2023, urging users to update to version 11.11.0.0. The flaw could be chained with CVE-2023-35081 to allow attackers to write malicious web shell files. Though there’s no detail on real-world exploits, federal agencies are advised to apply fixes by February 8, 2024. Two zero-day flaws in Ivanti Connect Secure VPN devices are also under mass exploitation, prompting the company to release updates next week. The attacks, initially linked to a Chinese threat actor, have since attracted additional threat actors globally, compromising over 2,100 devices across various sectors. Organizations are urged to apply Ivanti’s provided mitigation after importing backup configurations to prevent re-compromise.