Researchers uncovered critical security vulnerabilities in Azure DevOps, primarily focusing on Server-Side Request Forgery (SSRF) flaws. These weaknesses allowed unauthorized access to internal services, posing severe risks to the security of cloud-based development environments. During testing, the researcher identified issues in Service Connections that utilized privileged Service Principals, raising concerns about potential privilege escalation if attackers gained access to developer accounts. The vulnerabilities were discovered while analyzing endpoints related to creating Service Connections, revealing significant gaps in Azure DevOps’ security measures.
One of the key vulnerabilities was found in the endpointproxy API, which enabled outbound requests to internal services using valid Azure tokens. This discovery highlighted issues with token scope management within Azure Resource Manager (ARM), emphasizing the need for stricter security controls. The researcher also demonstrated the ability to bypass Microsoft’s earlier fixes for SSRF vulnerabilities through a DNS rebinding technique. By registering a domain that resolved to different IP addresses, they managed to access protected metadata APIs, showcasing the persistence of exploitable flaws in the system.
The researcher identified further SSRF vulnerabilities in Azure DevOps’ Service Hooks
Which failed to validate URL parameters properly. This allowed unauthorized access to external and internal services, creating opportunities for malicious exploitation. These findings were reported to Microsoft, which acknowledged the issues and awarded bounties for the discoveries. Despite these fixes, the researcher’s ability to uncover multiple vulnerabilities in a short timeframe suggests that other unaddressed flaws may still exist within the system.
This series of discoveries underscores the critical importance of maintaining robust security practices in cloud applications like Azure DevOps. Organizations using the platform should adopt comprehensive measures to safeguard sensitive data and prevent privilege escalation. The findings also highlight the evolving nature of cloud security threats, particularly in APIs and access controls, as cloud environments continue to grow in complexity. Ensuring strong protections against SSRF attacks and other vulnerabilities will be essential to maintaining trust and protecting user data in cloud-based systems.