CVE-2026-27727: mchange-commons-java: Remote Code Execution via JNDI Reference Resolution

Vulnerability summary

CVE-2026-27727 is a critical (CVSS 9.8) vulnerability in mchange mchange_commons_java. EPSS estimates a 0.98% chance of exploitation in the next 30 days, higher than 62.2% of scored vulnerabilities. 1 public exploit is recorded.

Timeline

1. 25-02-2026
   PublishedVulnerability first disclosed.
2. 15-09-2026
   Last modifiedVulnerability information updated.

Description

mchange-commons-java, a library that provides Java utilities, includes code that mirrors early implementations of JNDI functionality, including support for remote `factoryClassLocation` values, by which code can be downloaded and invoked within a running application. If an attacker can provoke an application to read a maliciously crafted `jaxax.naming.Reference` or serialized object, they can provoke the download and execution of malicious code. Implementations of this functionality within the JDK were disabled by default behind a System property that defaults to `false`, `com.sun.jndi.ldap.object.trustURLCodebase`. However, since mchange-commons-java includes an independent implementation of JNDI derefencing, libraries (such as c3p0) that resolve references via that implementation could be provoked to download and execute malicious code even after the JDK was hardened. Mirroring the JDK patch, mchange-commons-java's JNDI functionality is gated by configuration parameters that default to restrictive values starting in version 0.4.0. No known workarounds are available. Versions prior to 0.4.0 should be avoided on application CLASSPATHs.

• CWE-74Improper Neutralization of Special Elements in Output Used by a Downstream ComponentThe product constructs all or part of a command, data structure, or record using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify how it is parsed or interpreted when it is sent to a downstream component.
• CWE-502Deserialization of Untrusted DataThe product deserializes untrusted data without sufficiently ensuring that the resulting data will be valid.