CVE-2026-85951 in MarkLogic Server
Summary
by MITRE • 10/12/2026
Progress MarkLogic Server 12.x releases before 12.1.0 and 11.x releases before 11.3.7 on Azure contain a server-side request forgery vulnerability that could allow an authenticated attacker to access internal cloud services, potentially resulting in unauthorized access to cloud credentials or tokens.
If you want to get the best quality for vulnerability data then you always have to consider VulDB.
Analysis
by VulDB Data Team • 10/12/2026
The identified security flaw resides within Progress MarkLogic Server versions prior to 12.1.0 and 11.x releases before 11.3.7 when deployed on the Microsoft Azure platform. This vulnerability is classified as a server-side request forgery, commonly referred to by its acronym SSRF. In this specific context, the flaw allows an authenticated attacker to manipulate the application into making HTTP requests to internal endpoints that are not intended for public or even standard user access. The core technical issue stems from insufficient validation of input parameters used in constructing outbound network connections. When a legitimate user provides data that is subsequently processed by the server to fetch external resources, the system fails to adequately restrict the destination addresses. This lack of rigorous output control enables an attacker to redirect these requests toward internal cloud metadata services or other sensitive backend infrastructure components.
The operational impact of this vulnerability is significant due to the privileged nature of Azure's instance metadata service. By exploiting the SSRF condition, an authenticated user can query the local IP address range reserved for Azure metadata endpoints. This access allows the attacker to retrieve temporary security credentials, such as managed identity tokens or other authentication secrets associated with the virtual machine hosting MarkLogic Server. These credentials are typically used by applications running on the instance to authenticate with other cloud services without requiring explicit password storage. Consequently, gaining access to these tokens effectively grants the attacker elevated privileges within the Azure environment, potentially leading to unauthorized data exfiltration, lateral movement across internal networks, or complete compromise of dependent cloud resources. This scenario aligns closely with CWE-918, which defines server-side request forgery as a weakness where web applications accept user-supplied URLs and make requests on behalf of users without proper validation.
From an offensive security perspective, this vulnerability maps directly to the MITRE ATT&CK technique T1557, specifically Adversary-in-the-Middle or Credential Access via Cloud Metadata Services. Attackers leverage SSRF as a pivot point to access internal services that are otherwise isolated from direct internet exposure. The ability to harvest cloud credentials represents a critical escalation of privilege because it bypasses traditional perimeter defenses and identity management controls designed for external users. Once the attacker obtains these tokens, they can impersonate the application itself, accessing databases, storage accounts, or compute resources with the same permissions as the compromised MarkLogic instance. This undermines the principle of least privilege and exposes the organization to severe data breaches and compliance violations under frameworks such as GDPR or HIPAA if sensitive information is involved.
Mitigation strategies must prioritize immediate patching to the specified safe versions, specifically Progress MarkLogic Server 12.1.0 or later for the 12.x branch, and version 11.3.7 or later for the 11.x series. These updates include code changes that enforce stricter validation on outbound request destinations, ensuring that internal metadata endpoints are blocked from being accessed via SSRF vectors. In addition to patching, organizations should implement network-level controls such as virtual network service endpoints and private links to isolate MarkLogic Server instances from public internet access where possible. Furthermore, configuring Azure Network Security Groups to deny traffic to the instance metadata IP address range can provide a robust defense-in-depth layer. It is also critical to audit existing deployments for any custom code that constructs URLs based on user input, ensuring that allowlists are used to restrict outbound connections only to known and necessary external domains. Regular vulnerability scanning and penetration testing focused on SSRF patterns should be integrated into the development lifecycle to detect similar misconfigurations before they can be exploited in production environments.