Vulnerabilities

CVE-2026-114938 Java4Shell Vulnerability Enables Pre-Auth RCE in Keycloak, Jira, Jenkins and Other Enterprise Java Apps

November 18, 2026 08:01 · 8 min read
CVE-2026-114938 Java4Shell Vulnerability Enables Pre-Auth RCE in Keycloak, Jira, Jenkins and Other Enterprise Java Apps

Critical OpenJDK Base64 Decoder Flaw Exposes Enterprise Java Applications

Oracle and the OpenJDK project have released patches for a critical Java runtime vulnerability dubbed Java4Shell and tracked as CVE-2026-114938. The flaw resides in the way OpenJDK processes Base64-encoded data on x86-64 CPUs that support the AVX-512 VBMI instruction set. Security firm Calif disclosed the vulnerability on November 17 after details surfaced ahead of the coordinated disclosure timeline. Researcher Quang Luong discovered the bug, which Calif says took approximately one hour to find and two additional hours to develop a working exploit, with total AI token costs under $200.

Technical Root Cause: Heap Buffer Overflow in AVX-512 VBMI Decoding Routine

The vulnerability exists in a routine OpenJDK relies on to decode Base64 on processors supporting AVX-512 VBMI, including Intel Xeon processors from the Ice Lake generation onward and AMD chips based on Zen 4 or newer architectures. Systems powered by ARM64 chips are not affected. Calif's analysis demonstrates that invalid Base64 strings can cause the decoder to write past the end of its output buffer, corrupting adjacent heap data before Java flags the input as invalid.

This memory corruption capability has been named JavaBleed by the researchers. "With careful exploitation, the corruption can reveal sensitive memory, as Heartbleed did, or lead to remote code execution," Calif explained. Heap data exposed through this mechanism could contain personally identifiable information (PII), secret keys, and authentication tokens.

Pre-Authentication Remote Code Execution Achieved Against Major Enterprise Platforms

In initial testing, Calif achieved pre-authentication remote code execution (RCE) on the following applications running with default configurations:

Authenticated attackers could also execute code on Apache Kafka Broker, Solr, Elasticsearch, Gerrit, and Spinnaker. Calif emphasized that neither list is exhaustive. Code execution was additionally demonstrated against sample applications built on Apache Struts and Spring Boot.

Attack Vectors: Authentication, File Uploads, and Client-Side Scenarios

Calif warns that the vulnerable decoder can be reached through everyday functionality, including authentication workflows and file upload mechanisms. Client-side attacks are also feasible. Based on an exploit shared by security researcher samczsun, a Java application can be compromised when it connects to an attacker-controlled server over TLS, even if the application does not expose any network services.

"One thing to note about this vulnerability is that there is no universal exploit. Depending on how Base64 is used and how the application works, different steps may be required for exploitation," Calif noted.

Exploitation Complexity: DoS Trivial, RCE Requires Application-Specific Chains

Security firm VulnCheck confirmed that crashing a vulnerable JVM is straightforward once an attacker identifies a reachable entry point using the affected decoding logic. In their tests, a quick series of crafted requests reliably caused denial-of-service conditions. VulnCheck has also developed a full RCE exploit but noted that achieving code execution is significantly harder, requiring manipulation of heap layout and application-specific techniques to leak addresses and bypass ASLR.

"In other words, we don't anticipate that CVE-2026-114938 will be easily or widely exploitable, but as with PHP2Shell, adversaries are likely to find narrow but functional paths to RCE in specific applications," VulnCheck said.

The proof-of-concept code published by Calif demonstrates memory corruption but does not achieve code execution. Calif stated that weaponized exploitation chains for individual products will remain private until users have had time to upgrade.

Affected Versions and Vendor Patches

According to Calif, OpenJDK versions starting with 17.0.6 are vulnerable, along with downstream distributions based on them:

The OpenJDK advisory lists Java 17, 21, 25, and 27 as affected by CVE-2026-114938. Oracle released fixes in its November 2026 Critical Security Patch Update, which addresses 11 Java SE vulnerabilities. Oracle assigns the flaw a CVSS 3.1 score of 8.1, classifying it as remotely exploitable without authentication but with high attack complexity. Calif assigned a CVSS 4.0 score of 9.2.

Distribution Delays and CISA Urgency

Endor Labs pointed out that users of non-Oracle Java distributions must wait for their respective vendors to release patched builds. As of November 17, the fix was not yet present in the public OpenJDK 17, 21, and 25 update repositories. CISA has called for immediate patching, prioritizing internet-exposed systems. The agency also advises organizations to hunt for Java installations embedded in third-party software, containers, and appliances, warning these may not appear separately in asset inventories.

Threat Landscape and Detection

Cisco Talos reports no observed exploitation in the wild as of the disclosure date but noted: "we anticipate probing attempts may pick up soon after public disclosure." Organizations should prioritize inventory of all Java runtimes, apply vendor patches as they become available, and monitor for unusual JVM crashes or memory corruption indicators that could signal exploitation attempts.


Source: SecurityWeek

Source: SecurityWeek

Powered by ZeroBot

Protect your website from bots, scrapers, and automated threats.

Try ZeroBot Free