CVE-2021-39144: XStream is vulnerable to a Remote Command Execution attack.

Affected Versions

All versions until and including version 1.4.17 are affected, if using the version out of the box. No user is affected, who followed the recommendation to setup XStream's security framework with a whitelist limited to the minimal required types.


The processed stream at unmarshalling time contains type information to recreate the formerly written objects. XStream creates therefore new instances based on these type information. An attacker can manipulate the processed input stream and replace or inject objects, that result in execution of a local command on the server.

Steps to Reproduce

Create a simple PriorityQueue and use XStream to marshal it to XML. Replace the XML with following snippet and unmarshal it again with XStream:

<java.util.PriorityQueue serialization='custom'>
      <handler class='sun.tracing.NullProvider'>
              <proxy class='java.lang.Runtime'/>
XStream xstream = new XStream();

As soon as the XML gets unmarshalled, the payload gets executed and the command is executed on the host.

Note, this example uses XML, but the attack can be performed for any supported format. e.g. JSON.


The vulnerability may allow a remote attacker to execute commands with the rights of the process owner on the host only by manipulating the processed input stream.


See workarounds for the different versions covering all CVEs.


Ceclin and YXXX from the Tencent Security Response Center found and reported the issue to XStream and provided the required information to reproduce it.