Malware & Threats

Security Researchers Publish Java Flaw for Peer Review

Security Explorations, a start-up from Poland credited with the discovery of several Java flaws, has turned one of their discoveries out to the public for peer review. The problem is the fact that the researchers believe Oracle is wrongly classifying a legit security issue as expected behavior, so they want additional insight.

<p><span><span><strong>Security Explorations</strong>, a start-up from Poland credited with the discovery of several Java flaws, has turned one of their discoveries out to the public for peer review. The problem is the fact that the researchers believe Oracle is wrongly classifying a legit security issue as expected behavior, so they want additional insight. </span></span></p>

Security Explorations, a start-up from Poland credited with the discovery of several Java flaws, has turned one of their discoveries out to the public for peer review. The problem is the fact that the researchers believe Oracle is wrongly classifying a legit security issue as expected behavior, so they want additional insight.

The researchers published a document outlining Issue 54, which Oracle says isn’t an issue at all, as the process outlined as vulnerable is “allowed behavior.” Issue 54, the firm explains, stems from the fact that certain ‘method handle’ lookup methods are not invoking Java’s internal security checks (‘checkSecurityManager’).

When presented with the proof of concept, Oracle said “obtaining a method handle for a protected method from a superclass is allowed behavior.”

“Security Explorations didn’t agree with the [assessment] and on the same day provided its counterarguments to Oracle. We indicated that Issue 54 abused the missing security manager check [method] in order to gain access to method handle objects of certain security sensitive classes such as Class Loaders,” the firm explains (PDF).

In fact, when Issue 54 is combined with Issue 55 – a separate vulnerability discovered by the firm, a complete Java security sandbox bypass in the target system is possible.

Before publishing their work publically for peer review, Security Explorations waited weeks for Oracle to confirm their follow-ups and additional comments.

“Security Explorations believes that 3 weeks (from Feb 25 to Mar 18) constitutes enough time for a major software vendor to deliver a final confirmation or denial of a reported issue. This especially concerns a vendor that has been a subject of a considerable criticism regarding competent and prompt handling of security vulnerabilities in its software.”

Comments can be delivered to Security Explorations directly; the published details of Issue 54 are linked above. Oracle has yet to comment on the issue, or respond to the researchers.

Advertisement. Scroll to continue reading.

Related Content

Copyright © 2024 SecurityWeek ®, a Wired Business Media Publication. All Rights Reserved.

Exit mobile version