IBM UrbanCode Deploy could allow a user to execute code using a specially crafted file upload that would replace code on the server. This code could be executed on the UCD agent machines that host customer's production applications.
The agents in IBM UrbanCode Deploy 6.x before 6.0.1.14, 6.1.x before 6.1.3.3, and 6.2.x before 6.2.1.1 do not verify a server's identity in a JMS session or an HTTP session, which allows local users to obtain root access to arbitrary agents via unspecified vectors.
IBM UrbanCode Deploy 6.0.x before 6.0.1.13, 6.1.x before 6.1.3.3, and 6.2.x before 6.2.1.1, when agent-relay Codestation artifact caching is enabled, allows remote attackers to bypass authentication and obtain sensitive artifact information via unspecified vectors.
IBM UrbanCode Deploy 6.0.x before 6.0.1.13, 6.1.x before 6.1.3.3, and 6.2.x before 6.2.1.1 does not properly implement a logging-obfuscation feature for secure properties, which allows remote authenticated users to obtain sensitive information via vectors involving special characters.
IBM UrbanCode Deploy 6.0.x before 6.0.1.13, 6.1.x before 6.1.3.3, and 6.2.x before 6.2.1.1 allows remote authenticated users to obtain sensitive cleartext secure-property information via (1) the server UI or (2) a database request.
Multiple cross-site scripting (XSS) vulnerabilities in IBM UrbanCode Deploy 6.0 before 6.0.1.12, 6.1 before 6.1.3.2, and 6.2 before 6.2.0.2 allow remote authenticated users to inject arbitrary web script or HTML via a crafted URL.
IBM UrbanCode Deploy 6.0 and 6.0.1.x before 6.0.1.10, 6.1.1.x before 6.1.1.8, and 6.1.2 writes admin AUTH_TOKEN values to execution logs, which allows remote authenticated users to gain privileges by leveraging the ability to create and execute a process.