Critical CosmosEscape Flaw Breaks Customer Isolation in Azure Cosmos DB
A critical vulnerability dubbed CosmosEscape could have allowed attackers to take control of virtually every instance of Azure Cosmos DB, including Microsoft’s own internal databases. The flaw broke the isolation boundary between different customers on the cloud platform.
The issue resided in the Gremlin API, which is used for graph-oriented database queries. Carefully crafted queries escaped the restricted sandbox and executed code on the infrastructure responsible for processing customer requests. Researchers chained the exploit by abusing limitations in .NET reflection protections.
Once inside the DB Gateway, the attackers developed techniques to read and write files and run arbitrary commands. Successful compromise of the gateway granted access to a global signing secret known as the Cosmos Master Key. This key enabled retrieval of the primary key for any account, providing complete read and write access.
The attack surface also included the Config Store, a directory holding account names, subscription and tenant identifiers, network rules, and other configuration data. Attackers could filter this information to target specific organizations. Even privately isolated instances remained exposed because the DB Gateway itself enforced network restrictions.
Internal Microsoft services such as Microsoft Entra ID, Teams, and Copilot rely on Cosmos DB, underscoring the broad potential reach of the vulnerability.
Related articles
Dangerous C++ Traps: Memory Safety Issues, Undefined Behavior, and Code That Betrays Developers
Around 70% of vulnerabilities assigned CVE numbers by Microsoft each year stem from memory safety errors, with Chromium reporting a similar pattern for serious Chrome bugs. The article examines how C++ deliberately permits low-level memory access for performance but demands strict discipline to avoid undefined behavior (UB). It details real-world cases including Heartbleed (CVE-2014-0160) in OpenSSL, use-after-free errors, buffer overflows, and data races. Modern tools such as AddressSanitizer, UndefinedBehaviorSanitizer, and ThreadSanitizer are presented as essential for detection, alongside recommendations for RAII, std::span, and smart pointers. The piece also discusses C++26 changes introducing erroneous behavior for uninitialized variables and advises when to consider memory-safe languages like Rust for new components. Practical migration steps and compiler warning strategies are outlined to reduce risk in existing codebases.
Critical RCE Vulnerability CVE-2026-66066 Affects Ruby on Rails Active Storage with libvips
A severe vulnerability tracked as CVE-2026-66066 has been identified in the Ruby on Rails web application framework. The flaw, also referred to as KindaRails2Shell by researchers, impacts applications that use Active Storage with the libvips image processing engine. Attackers can exploit the issue by uploading specially crafted files to read arbitrary files without authentication. This exposure may lead to theft of environment variables, secret keys, and external service credentials, enabling remote code execution or further attacks. The vulnerability stems from unsafe operations in the dependent libvips library. It carries a CVSS v4.0 base score of 9.5 and is rated Critical. JPCERT/CC has issued an advisory urging immediate updates.
Critical Vulnerabilities Disclosed in Adobe Campaign Classic Require Immediate Patching
Adobe has released a security advisory detailing two high-risk vulnerabilities affecting Adobe Campaign Classic on-premises deployments on Windows and Linux. CVE-2026-48449 is an improper authorization flaw that permits remote attackers to execute arbitrary code without authentication and carries a maximum CVSSv3.1 base score of 10.0. CVE-2026-48448 is a SQL injection vulnerability that allows unauthenticated attackers to read arbitrary files from the file system, rated at 8.6. Both issues affect the campaign management platform used by organizations for marketing automation. Adobe urges administrators to apply the available security updates without delay due to the elevated risk of exploitation.
Secure Error Logging Practices to Prevent Information Leaks Across Java, Kotlin, JavaScript and Python
The article examines how improper error logging can expose sensitive details such as stack traces, file paths, library versions and database structures, enabling attackers to map applications and craft targeted exploits. It covers core logging levels from DEBUG to FATAL, mechanisms including text files, binary logs and databases, plus the roles of Trace ID and Correlation ID in tracing requests across microservices. Real-world vulnerable code examples in Flask and SQLite demonstrate direct exception output, manual traceback exposure and SQL error leakage that confirm technologies like Python 3.10 or SQLite usage. CWE categories including CWE-209, CWE-532, CWE-538 and CWE-1295 are referenced to classify risks of information disclosure through logs. Mitigation steps include stripping version headers in Nginx, sanitizing inputs with regular expressions, using OpenTelemetry for structured JSON logging and avoiding debug modes in production. The guidance stresses balancing detailed logs for incident response with protections against injection and reconnaissance.