top of page

Oracle Shipped 11 Java Fixes Today. The Reported OpenJDK RCE, CVE-2026-114938 (Java4Shell), Is Still Unconfirmed. Which Builds to Move To, and Where Java Hides.

Writer: Patrick Duggan
Patrick Duggan
50 minutes ago
7 min read

If you run Oracle's JDK, move to the November 17 builds today: 8u513, 11.0.33.1, 17.0.21.1, 21.0.13.1, 25.0.5.1 or 27.0.1.1. Oracle has posted release notes for all six, each dated November 17, 2026. If your Java comes from someone else (Eclipse Temurin, Amazon Corretto, Azul Zulu, Red Hat, the Microsoft Build of OpenJDK, your Linux distribution, or a vendor that bundles its own runtime), Oracle's download does not patch you. Those vendors build their own binaries on their own schedules. Check your vendor's release page and install their November build when it lands.


Separately, a reported unauthenticated remote code execution flaw in OpenJDK, CVE-2026-114938, nicknamed "Java4Shell," is circulating. As of 19:31 UTC on November 17, no primary source we could read confirms its details: there is no OpenJDK advisory for it, no published CVE record, and Oracle has not named it. No exploitation in the wild has been reported, and it is not in CISA's Known Exploited Vulnerabilities catalog. Treat it as a reason to patch fast, not as a confirmed fact.



What to do, in order


  1. Find every Java runtime you have (list below). Most small shops run more copies than they think.

  2. For each one, note who built it. Run java -version with that copy's own binary. The vendor name is usually in the output (Oracle, Temurin, Corretto, Zulu, Red Hat, Microsoft).

  3. Oracle JDK: install the November 17 build for your line. The full version strings, from Oracle's release notes, are 1.8.0_513-b04, 11.0.33.1+4, 17.0.21.1+4, 21.0.13.1+4, 25.0.5.1+4 and 27.0.1.1+4.

  4. Any other vendor: watch their release page. At 19:31 UTC we checked two of them. The latest Eclipse Temurin builds in the Adoptium API were still the October and early-November ones (17.0.21+9 and 21.0.13+9 from October 22, 8u512 and 11.0.33+10 from November 5). The latest Amazon Corretto releases on GitHub were 17.0.21.0.11 and 21.0.13.0.11, both from October 27. Neither had a build newer than Oracle's announcement yet. That will change; check again rather than trusting this paragraph.

  5. Restart the process. A patched JDK on disk does nothing for a JVM that has been running since September.

  6. Prioritize Java that parses data from the internet: web apps, API servers, message consumers, anything behind a public load balancer.


Where Java hides in a small shop


Java is rarely installed on purpose and then forgotten. It arrives inside something else. Look in these places:


  • Application servers: Tomcat, WildFly/JBoss, WebLogic, WebSphere, Jetty. Check which JVM the service actually starts with (JAVA_HOME in the service file or startup script), not just which one is on the PATH.

  • Build and CI: Jenkins controllers and every Jenkins agent, plus TeamCity, Bamboo, and Gradle or Maven build hosts.

  • Search and data: Elasticsearch and OpenSearch ship a bundled JDK inside their own install directory (a folder named jdk). Upgrading the system Java does not touch it; you upgrade Elasticsearch or OpenSearch to get a new bundled runtime. Kafka, ZooKeeper, Cassandra, Solr, Hadoop and Spark all run on Java too.

  • Atlassian and friends: Jira, Confluence and Bitbucket Data Center ship with their own runtime in many installs.

  • Containers: base images such as eclipse-temurin, amazoncorretto and openjdk pin a Java build at build time. Rebuild the image; restarting the container is not enough.

  • Vendor appliances: backup software, network management consoles, VPN and firewall management servers, printer and badge systems, and lab instruments often bundle a private JRE (look for a folder named jre or jdk under the product's install path). You usually cannot patch these yourself. Ask the vendor when they will ship the November Java fixes.

  • Desktops: developer laptops (IntelliJ, Eclipse, Android Studio ship their own runtimes), Minecraft servers, and old line-of-business apps.

A quick hunt on Linux: search the filesystem for files named java under bin directories, then run each one with -version. On Windows, search Program Files for java.exe and check the registry and "Apps and features" for every Java entry. Oracle's own Java Management Service can inventory Java installs if you are an Oracle customer.



What is confirmed


From Oracle's Critical Security Patch Update Pre-Release Announcement for November 2026 (the announcement page, which was still the pre-release version when we read it at 19:31 UTC):


  • The update "will be released on Tuesday, November 17, 2026" and "addresses 652 new security patches." That 652 is the total across all Oracle products in this update (Database, Fusion Middleware, E-Business Suite, Hyperion and more), not Java.

  • For Java: "This Critical Security Patch Update contains 11 new security patches for Oracle Java SE. All of these vulnerabilities may be remotely exploitable without authentication, i.e., may be exploited over a network without requiring user credentials."

  • "The highest CVSS v3.1 Base Score of vulnerabilities affecting Oracle Java SE is 8.1." That is the top score among those 11 Java SE fixes.

  • Affected versions listed: Oracle Java SE 8u511, 8u511-perf, 11.0.33, 17.0.21, 21.0.13, 25.0.5, 27.0.1; Oracle GraalVM for JDK 17.0.21 and 21.0.13; Oracle GraalVM Enterprise Edition 21.3.20.0.3; Oracle JavaFX 21.0.13, 25.0.5, 27.0.1.

From Oracle's JDK release notes: release notes for JDK 8u513, 11.0.33.1, 17.0.21.1, 21.0.13.1, 25.0.5.1 and 27.0.1.1 are posted. The 8, 11, 17, 21 and 25 notes carry "Release date: November 17, 2026." The 21.0.13.1 notes say "This release contains fixes for security vulnerabilities described in the Oracle Critical Patch Update," and list the new JRE security baselines: 21.0.13.1+4, 17.0.21.1+4, 11.0.33.1+4 and 1.8.0_513-b04. Oracle also adds one behavior change in these builds: a Kerberos cross-realm transited-field check is now on by default (it can be disabled with the sun.security.krb5.acceptor.checkTransitedPolicy property, which Oracle says to do "only after carefully considering the security implications"). If you run Kerberos with Java, test that first.


Two caveats from the announcement itself. Oracle says the pre-release information "may change before publication of the Critical Security Patch Update Advisory." And the announcement does not give CVE IDs, components, or whether any of the 11 needs untrusted code (a sandboxed applet or Web Start style deployment) versus data sent to a server API. Oracle normally puts that in the risk matrix of the full advisory, which had not replaced the announcement when we read it.



What is not confirmed: [CVE-2026-114938](https://analytics.dugganusa.com/api/v1/dredd/kev-gap?cve=CVE-2026-114938), "Java4Shell"


What is being said:


  • Calif, a security research firm, reportedly says it found and reported the flaw, named it Java4Shell, and says the details "leaked." The Hacker News reports Calif said it would publish technical details "by 1 p.m. Pacific time on November 17" (21:00 UTC), after Oracle's patch. We checked Calif's blog feed at 19:31 UTC; no Java post was there yet. Calif's post on X returned an error to our fetch, so we could not read it directly.

  • Several vendors (Aikido, watchTowr, Datadog, Rapid7) relay the same summary: unauthenticated remote code execution in OpenJDK 17.0.6 and later, CVSS 4.0 score 9.2. Rapid7 says plainly: "These CVE-specific details remain preliminary, pending confirmation in Oracle or OpenJDK advisories." These are several outlets, but as far as we can trace they lead back to one origin, the Calif report. Read them as one source, not five.

  • One news aggregator attributes a root cause to Oligo Security: an integer underflow in OpenJDK's AVX-512 Base64 decoder, triggered by a crafted 65-byte input. We read Oligo's own blog post. It says the flaw is "a critical memory corruption vulnerability in Java's core" on x86 systems, but it does not mention Base64, AVX-512 or the 65-byte input. We could not find that detail in any primary source, so we are not passing on the JVM-flag workarounds tied to it. Changing intrinsic flags on a production JVM because an aggregator said so is not a plan.

What is checked and missing:


  • The OpenJDK Vulnerability Group's advisory list, at 19:31 UTC, showed no November advisory. Its latest is dated 2026/10/24 (an out-of-cycle fix for a different flaw, CVE-2026-114820, CVSS 3.1 9.1 in security-libs/java.security, affecting all of 8 through 27).

  • The CVE.org API returns "CVE_RECORD_DNE" for CVE-2026-114938: the ID has no published record. The Hacker News describes it as reserved.

  • Oracle's November announcement does not mention it, and The Hacker News reports "no evidence yet" that it is among Oracle's 11 Java fixes.

  • CISA's KEV catalog JSON, downloaded at 19:31 UTC, does not contain it.

  • We found no public proof-of-concept and no report of exploitation.

So the honest state is: Oracle's 11 Java fixes are real and shipped. Java4Shell may or may not be one of them. Note too that the reported version range (17.0.6 and later) is OpenJDK, which means every vendor build in that range is in question, not just Oracle's. If the report holds, the fix that protects a Temurin or Corretto user is a Temurin or Corretto build, and those had not shipped when we checked.



Indicators


None. No primary source published network indicators or file hashes, so we added nothing to our feed.



How we know


We downloaded and read Oracle's November 2026 pre-release announcement and the release notes for each JDK line from oracle.com, the OpenJDK advisory index and its 2026/10/24 advisory, the CVE.org record API, the CISA KEV JSON, the Adoptium release API, and Corretto's GitHub releases, all between 19:28 and 19:31 UTC on November 17. We read The Hacker News, Datadog Security Labs, Rapid7 and Oligo for the Java4Shell claims. Our edge check for this CVE could not run (our scanner host timed out), and there is no published exploit path to match against anyway, so we make no claim about what our edge has or has not seen.



Sources


  • Oracle, Critical Security Patch Update Pre-Release Announcement, November 2026: https://www.oracle.com/security-alerts/cspunov2026.html

  • Oracle, JDK 21.0.13.1 release notes: https://www.oracle.com/java/technologies/javase/21-0-13-1-relnotes.html (also 8u513, 11.0.33.1, 17.0.21.1, 25.0.5.1 and 27.0.1.1 at the same path pattern)

  • OpenJDK Vulnerability Advisories: https://openjdk.org/groups/vulnerability/advisories/

  • OpenJDK advisory 2026/10/24: https://openjdk.org/groups/vulnerability/advisories/2026-10-24

  • CVE record (not yet published): https://www.cve.org/CVERecord?id=CVE-2026-114938

  • The Hacker News, "Oracle to Patch 11 Java Flaws Today as Critical OpenJDK RCE (CVE-2026-114938) Report Emerges": https://thehackernews.com/2026/11/oracle-to-patch-11-java-flaws-today-as.html

  • Rapid7: https://www.rapid7.com/blog/post/etr-cve-2026-114938-reported-critical-pre-authentication-remote-code-execution-java-17

  • Datadog Security Labs: https://securitylabs.datadoghq.com/articles/java4shell-remote-code-execution-in-openjdk-cve-2026-114938

  • Oligo Security: https://www.oligo.security/blog/java4shell-oligo-blocks-the-zero-day-before-the-patch

  • Calif blog (no Java post at time of reading): https://blog.calif.io/

  • Eclipse Temurin releases: https://adoptium.net/temurin/releases/

  • Amazon Corretto releases: https://github.com/corretto



Every indicator in this post is in the feed. Free.

1.58M+ IOCs, STIX 2.1 / TAXII, 88% novel vs ThreatFox, exploited-CVE leads ahead of CISA. No credit card — a free API key in 30 seconds, and you can audit every claim above against the live endpoints.


Comments

Rated 0 out of 5 stars.
No ratings yet

Add a rating
bottom of page