HIGH 7.5

CVE-2026-13772: IBM WebSphere Extreme Scale OQL Remote Code Execution Vulnerability

IBM WebSphere Extreme Scale versions 8.6.1.0 through 8.6.1.6 contain a critical vulnerability in their Object Query Language (OQL) engine that allows authenticated attackers to execute arbitrary code on affected application servers. The vulnerability stems from unsafe handling of class names in OQL queries—the system dynamically loads and instantiates classes based on user input without any validation or blocklist. An attacker with valid credentials who can influence OQL query strings used by an application can trigger execution of arbitrary constructors within the WebSphere JVM. Additionally, a serialization variant can bypass Java's standard security filters when data moves between grid nodes, expanding the attack surface.

Source data · NVD / CISA · public domain

CVSS
3.1 · 7.5 HIGH · CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H
Weaknesses (CWE)
CWE-470
Affected products
1 configuration(s)
Published / Modified
2026-06-30 / 2026-07-03

NVD description (verbatim)

IBM WebSphere Extreme Scale 8.6.1.0 through 8.6.1.6 's Object Query Language engine resolves attacker-supplied class names via Class.forName() and invokes their constructors with no allow-list at three distinct sinks (SELECT NEW, enum literals, and reflection-based comparators); an authenticated remote attacker who can influence an application-built OQL query string can execute arbitrary constructors on the WAS JVM, and a SELECT DISTINCT variant using planted grid values fires the same gadget post-readObject in a manner that survives JEP-290 serialization filters across grid node boundaries

1 reference(s) · View on NVD →

SEC.co analysis · AI-assisted, reviewed against source

Technical summary

The vulnerability arises from three distinct code paths in the OQL engine that invoke Class.forName() on attacker-controlled input without an allow-list: SELECT NEW clauses, enum literal resolution, and reflection-based comparators. An authenticated attacker controlling OQL query parameters can instantiate arbitrary classes, achieving remote code execution via gadget chain activation. A secondary attack vector involves SELECT DISTINCT queries that plant malicious grid values; these trigger the same gadget execution during deserialization (readObject()) in a manner that circumvents JEP-290 serialization filters across distributed grid node boundaries. The lack of input validation at the OQL parsing stage is the root cause.

Business impact

Successful exploitation grants an authenticated remote attacker code execution privileges on WebSphere Extreme Scale JVMs, potentially compromising in-memory data grids, cached application data, and the underlying infrastructure. Organizations using WebSphere Extreme Scale in production for distributed caching, session management, or data grid services face risk of data exfiltration, data corruption, lateral movement into dependent systems, and denial of service. The authenticated requirement narrows the immediate attack surface, but insider threats and compromised application credentials remain realistic vectors.

Affected systems

IBM WebSphere Extreme Scale versions 8.6.1.0 through 8.6.1.6 are affected. Organizations should inventory deployments of these versions, particularly those exposed to untrusted networks or supporting multi-tenant applications where query strings might be influenced by application users. Earlier and later versions should be verified against IBM's official security advisories to confirm patch status.

Exploitability

Exploitation requires valid authentication credentials and the ability to influence OQL query construction within the application. The CVSS score of 7.5 (HIGH) reflects authentication as a prerequisite (PR:L) and locally constrained attack complexity (AC:H), yet the impact is severe—confidentiality, integrity, and availability are all fully compromised. The serialization bypass variant may lower practical barriers if applications exchange grid data across nodes without proper isolation. No public exploit code is known at this time, but the vulnerability is not listed on the CISA Known Exploited Vulnerabilities catalog, indicating limited active exploitation visibility to date.

Remediation

Apply IBM security patches to upgrade WebSphere Extreme Scale to a patched version; consult IBM's official security advisory for exact version numbers and delivery timelines. As an interim measure, restrict OQL query construction—avoid dynamic query building from user input, implement strict parameterized queries, and enforce network-level access controls limiting OQL exposure to trusted hosts only. Disable or isolate grid node communication where feasible pending patch deployment.

Patch guidance

IBM will issue patches to address this vulnerability; verify exact patched versions and release dates against the official IBM WebSphere Extreme Scale security advisory. Patch deployment should prioritize production environments handling sensitive data or supporting critical applications. Test patches in a staging environment replicating your grid topology before production rollout, as OQL-dependent applications must be validated post-patching.

Detection guidance

Monitor WebSphere Extreme Scale logs and network traffic for anomalous OQL queries containing suspicious class names (particularly those invoking constructors for non-standard Java classes or gadget chain components). Inspect application code to identify dynamic OQL query construction; flag instances where user input flows into SELECT NEW, enum resolution, or comparator logic. Search for SELECT DISTINCT queries with unusual grid value patterns. Network-level detection should flag unexpected serialization patterns or object streams crossing grid boundaries.

Why prioritize this

This vulnerability merits urgent attention despite the authentication requirement: it enables remote code execution on a critical middleware component (in-memory data grid), affects multiple attack vectors (OQL clauses and serialization), and bypasses standard Java security controls (JEP-290). Organizations relying on WebSphere Extreme Scale for session storage, distributed caching, or real-time data services should patch immediately. The HIGH severity rating and full impact (confidentiality, integrity, availability) justify expedited remediation.

Risk score, explained

CVSS 7.5 reflects a HIGH severity vulnerability with remote network-level attack vector (AV:N), authentication requirement (PR:L), and high attack complexity (AC:H). The score appropriately weights the severe impact (C:H, I:H, A:H) against the credential prerequisite. However, context matters: organizations with strict application-layer query controls, network isolation, or no WebSphere Extreme Scale deployment face lower practical risk; those with dynamic query patterns or untrusted application tenants should treat this as CRITICAL.

Frequently asked questions

Can an unauthenticated attacker exploit this vulnerability?

No. The vulnerability requires valid authentication credentials (PR:L in the CVSS vector). An attacker must have legitimate access to the application or WebSphere Extreme Scale instance. However, compromised credentials, insider threats, or privilege escalation from other vulnerabilities could provide the necessary foothold.

Does the serialization bypass bypass all Java security filters?

The SELECT DISTINCT variant circumvents JEP-290 serialization filters specifically when data crosses grid node boundaries. Standard deserialization protections on a single node may still apply; the vulnerability is most critical in distributed deployments where grid nodes exchange serialized objects.

Is there a workaround if I cannot patch immediately?

Yes, implement application-level mitigations: avoid dynamic OQL query construction from user input, use parameterized query patterns, restrict network access to OQL endpoints to trusted internal hosts, and monitor query logs for suspicious class instantiation attempts. These are interim measures; patching remains the definitive fix.

Which versions are safe?

IBM WebSphere Extreme Scale 8.6.1.0 through 8.6.1.6 are vulnerable. Verify the safety of versions outside this range and any patched releases by consulting IBM's official security advisory and release notes.

This analysis is provided for informational purposes and reflects publicly available vulnerability data as of the publication date. Specific patch version numbers, release dates, and detailed IBM remediation steps must be verified against IBM's official security advisories. Organizations should conduct their own risk assessment based on deployment architecture, exposure, and compensating controls. No exploit code or weaponized proof-of-concept is provided. SEC.co makes no warranty regarding the completeness or accuracy of third-party vendor information. Source: NVD (public-domain), retrieved 2026-08-09. Analysis generated by SEC.co (claude-haiku-4-5).