Executive brief
Jenkins, a widely used automation server for software development, contains a critical security flaw that allows unauthorized individuals to take complete control of the server. By sending a specially crafted command to the Jenkins command-line interface, an attacker can execute arbitrary code without needing a username or password. This could lead to the theft of sensitive source code, disruption of software delivery pipelines, or the use of the server as a foothold for further attacks within the corporate network.
Technical details
A remote code execution vulnerability exists in Jenkins due to the improper deserialization of untrusted data within the Jenkins CLI. The vulnerability stems from the CLI's use of a remoting-based protocol that utilizes Java serialization. An unauthenticated attacker can bypass existing blocklist protections by transferring a serialized Java 'SignedObject' object, which is then deserialized using a new 'ObjectInputStream'. This bypass allows for arbitrary code execution on the Jenkins controller. The issue is addressed in Jenkins 2.57 and 2.46.2 LTS by adding 'SignedObject' to the blocklist and deprecating the remoting-based CLI protocol in favor of a new HTTP-based protocol.
Affected products
- Jenkins Jenkins 2.56 and earlier, 2.46.1 LTS and earlier
Timeline
- 2017-04-26: advisory: Original vendor advisory published by Jenkins
- 2025-10-02: kev added: Added to CISA Known Exploited Vulnerabilities catalog
- 2025-10-02: exploited: Confirmed as exploited in the wild per CISA KEV entry