October 6, 2025
Updated: September 2, 2026
A 2026 defensive guide to affected Oracle EBS versions, the current patch path, evidence-based hunting, incident response, and safe post-patch validation.
Mohammed Khalil

What changed in this update: This article now reflects Oracle’s October 2025 cumulative update guidance, the related CVE-2025-61884 alert, Google’s 2026 retrospective, CISA KEV status, evidence-based hunting, and safe post-patch validation. Operational exploit commands and undated IOC copying were removed.
CVE-2025-61882 is an unauthenticated remote code execution vulnerability in Oracle E-Business Suite 12.2.3 through 12.2.14, affecting Oracle Concurrent Processing’s BI Publisher Integration. Oracle rates it 9.8 and issued an emergency update in October 2025; CISA added it to the Known Exploited Vulnerabilities catalog. In 2026, organizations should confirm the current Oracle-supported patch level, investigate historical exposure instead of treating installation as proof of cleanliness, restrict unnecessary EBS egress, and validate remediation safely. Never use public exploit code against production or treat one IOC match as proof of compromise.
| Field | Confirmed information |
|---|---|
| Product | Oracle E-Business Suite Oracle Concurrent Processing |
| Component | BI Publisher Integration |
| Supported affected versions | 12.2.3 through 12.2.14 |
| Attack path | Network access through HTTP; HTTPS is also in scope under Oracle’s protocol convention |
| Authentication required | No |
| User interaction required | No |
| Impact | Remote code execution and potential takeover of Oracle Concurrent Processing |
| CVSS v3.1 | 9.8 Critical `CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H` |
| Exploitation status | In CISA KEV; known ransomware-campaign use |
| Original Oracle alert | October 4, 2025; revised October 6, 2025 |
Oracle’s CVE-2025-61882 Security Alert says successful exploitation can compromise Oracle Concurrent Processing and result in remote code execution. Oracle’s risk matrix describes an easily exploitable, unauthenticated network path with high confidentiality, integrity, and availability impact.
CISA added the flaw to its Known Exploited Vulnerabilities Catalog on October 6, 2025, assigned a federal remediation date of October 27, and marks known ransomware-campaign use as “Known.” The federal deadline has passed, but KEV status remains a strong prioritization signal for every organization running an affected EBS deployment.
This is a case where a mature vulnerability management lifecycle should override routine queue order. Confirm affected assets, exposure, and business criticality immediately; do not wait for a scanner’s generic severity workflow to catch up.
CVE-2025-61882 affects the BI Publisher Integration component within Oracle Concurrent Processing. In practical terms, a reachable attacker does not need a valid EBS username, password, or user click to reach the vulnerable path. If exploitation succeeds, attacker-controlled code can run in the context available to the affected EBS processing environment, creating a route to sensitive application data and follow-on activity.
Oracle has not publicly documented a complete root-cause analysis. The CVE record’s CISA enrichment maps the weakness to improper authentication, but defenders should avoid reducing the campaign to one public payload or one endpoint. Google/Mandiant observed multiple chains, and public reporting did not cleanly map every chain to one CVE.
The distinction between a vulnerability assessment and penetration testing matters here. Version and patch checks can identify likely exposure; a carefully authorized validation can test whether the known path is closed. Neither activity, by itself, determines whether historical compromise or data theft occurred.
| Date | Publicly reported development | Confidence note |
|---|---|---|
| July 10, 2025 | Suspicious traffic was observed against EBS environments | Google said the exact nature could not be confirmed |
| August 9, 2025 | Earliest assessed zero-day exploitation | Later review says CVE-2025-61882 and/or CVE-2025-61884 |
| September 29, 2025 | High-volume extortion emails began reaching executives | Sender claimed affiliation with the CL0P brand |
| October 4, 2025 | Oracle released the CVE-2025-61882 emergency alert | Alert addressed unauthenticated RCE risk |
| October 11, 2025 | Oracle released the CVE-2025-61884 alert | Separate EBS flaw; not interchangeable with 61882 |
| October 21, 2025 | Oracle published the October 2025 CPU | EBS CPU included both alerts plus additional fixes |
Google’s March 2026 zero-day review states that attackers exploited CVE-2025-61882 and/or CVE-2025-61884 as zero-days as early as August 9, with additional suspicious activity dating to July 10. That “and/or” is important: it is more precise than assigning every observed intrusion to CVE-2025-61882 alone.
The extortion activity used the CL0P name and contact infrastructure associated with that brand. However, Google Threat Intelligence Group did not formally attribute all activity to a tracked group and said the CL0P brand was not exclusive to FIN11. DeepStrike’s Cl0p ransomware profile provides broader campaign context, but attribution should not substitute for evidence from the affected environment.
These are related by timing and product family, but they are not the same vulnerability.
| Attribute | CVE-2025-61882 | CVE-2025-61884 |
|---|---|---|
| EBS component | Oracle Concurrent Processing BI Publisher Integration | Oracle Configurator Runtime UI |
| Supported affected versions | 12.2.3–12.2.14 | 12.2.3–12.2.14 |
| Authentication | None | None |
| CVSS v3.1 | 9.8 Critical | 7.5 High |
| Vendor-described impact | Takeover of Oracle Concurrent Processing; RCE | Access to sensitive resources |
| Alert date | October 4, 2025 | October 11, 2025 |
Do not mark an EBS environment complete because only one alert was addressed. The investigation and patch record should account for both alerts, the July 2025 update context, and any later cumulative updates applicable to the exact EBS release and platform.
The original CVE-2025-61882 alert says the October 2023 Critical Patch Update is a prerequisite for applying its emergency updates. That statement describes the original alert path; it is not permission to assume the same one-off sequence is the best 2026 route for every EBS system.
Oracle’s October 2025 Critical Patch Update says its EBS update includes fixes for CVE-2025-61882 and CVE-2025-61884 plus additional patches. That cumulative context is why current administrators should use the applicable Patch Availability Document and My Oracle Support rather than copying patch numbers or commands from an old blog post.
Oracle now publishes quarterly CPUs and, since 2026, additional Critical Security Patch Updates through its security advisory hub. A current CSUP is not automatically applicable to every EBS environment. Confirm product coverage, prerequisites, supersedence, platform-specific instructions, downtime, rollback, and post-install validation in Oracle’s current documentation.
A defensible patch management process for this issue should produce more than an “installed” status. Record the asset, EBS release, previous CPU level, selected Oracle guidance, patch identifiers, change window, installer output, restarts, functional tests, failed nodes, exceptions, rollback state, and final verification.
Unsupported or older releases should not be assumed safe merely because they fall outside Oracle’s supported-version table. Oracle says untested earlier releases are likely affected and recommends upgrading to a supported release.
A patch changes future exploitability; it does not rewrite history. If an EBS endpoint was reachable before the applicable update, the organization needs a compromise assessment covering the exposure window and the available evidence. That remains true when the patch installs cleanly and the current vulnerability scan is negative.
This distinction is central to the broader external penetration testing process. A present-day exposure test asks whether an attack path remains reachable. Incident response asks whether an attacker used that or another path earlier, what they did, what data they accessed, and whether persistence remains.
The evidence window should begin no later than July 10, 2025 when logs exist, because Google/Mandiant reported suspicious activity from that date and possible zero-day exploitation from August 9. Extend the review through patching, restarts, and the period needed to establish that no persistence, unauthorized access, or exfiltration continued.
Google/Mandiant’s original Oracle EBS campaign analysis documents anomalous application requests, database-stored template artifacts, Java-launched shell activity, outbound connections, in-memory components, and data theft in some cases. Use the maintained source for current technical indicators instead of copying a frozen list into a long-lived article.
An IOC is a lead, not a verdict. An IP address can be reassigned, a path can appear in benign administration or scanning, and a hash match may identify a downloaded proof-of-concept rather than a successful intrusion. Conversely, absence of known IOCs does not prove that a host was never compromised.
| Confidence level | Example evidence pattern | Recommended interpretation |
|---|---|---|
| Low | One historical source-IP or URI match with no corroboration | Preserve and enrich; do not declare compromise |
| Medium | Anomalous request plus unusual EBS template creation, errors, or correlated egress | Escalate to incident triage and collect volatile evidence |
| High | EBS Java process spawning an unexpected shell, confirmed attacker code execution, unauthorized database access, persistence, or staged data | Treat as a confirmed or highly probable incident and contain |
| Data-loss evidence | Unapproved large transfer, attacker-provided legitimate file listing, or verified access to sensitive records | Launch data-scope, legal, contractual, and notification analysis |
| Layer | Evidence to review | What it can help establish |
|---|---|---|
| Edge | Reverse proxy, load balancer, WAF, VPN, firewall, and DNS logs | Reachability, request timing, source context, and possible egress |
| EBS application | Web access, application, Concurrent Processing, BI Publisher, and Configurator logs | Targeted requests, errors, execution timing, and account context |
| Database | Audit records, template and LOB creation history, query history where enabled | Suspicious stored content and access to business data |
| Host | Process lineage, memory, file-system timeline, scheduled jobs, startup changes, and authentication logs | Execution, persistence, and attacker activity |
| Identity | EBS, OS, database, service-account, and privileged-access records | Credential use, privilege changes, and scope |
| Network and data | Egress, proxy, flow, DLP, storage, and backup telemetry | Command-and-control, staging, exfiltration, and recovery impact |
Preserve original timestamps, time-zone settings, retention limits, and chain of custody. If volatile evidence may matter, avoid unnecessary reboots or cleanup until responders decide how to capture it.
If evidence suggests exploitation, use the organization’s existing incident response plan and adapt it to the EBS environment. The sequence below is a decision aid, not a substitute for forensic, legal, Oracle Support, or business-owner direction.
| Window | Priority actions | Evidence and decision output |
|---|---|---|
| First 0–4 hours | Activate incident leadership; identify affected instances; preserve logs and volatile evidence; restrict unnecessary exposure and egress; avoid destructive cleanup | Named incident owner, asset list, preservation record, containment decision |
| 4–12 hours | Correlate edge, EBS, database, host, identity, and network evidence; apply approved Oracle updates to still-vulnerable systems without destroying needed evidence | Initial confidence assessment, patch record, suspected accounts and data paths |
| 12–24 hours | Contain confirmed execution; rotate exposed credentials and secrets based on evidence; involve Oracle Support, DFIR, legal, privacy, and communications as appropriate | Incident scope hypothesis, business-impact decision, next evidence plan |
| 24–72 hours | Complete forensic scoping, persistence review, data-access analysis, recovery validation, and enhanced monitoring | Defensible incident conclusion, recovery approval, notification analysis, lessons learned |
Isolation choices should account for EBS business criticality and evidence preservation. If immediate disconnection would create material operational harm, use risk-approved segmentation, inbound restrictions, egress controls, credential controls, and heightened monitoring while a coordinated containment plan is executed.
| Stage | Decision | Minimum evidence |
|---|---|---|
| P — Pinpoint | Which EBS assets, versions, nodes, clones, and routes are affected or exposed? | Inventory, topology, ownership, exposure record |
| A — Apply | Which current Oracle-supported update path is correct? | MOS guidance, change record, installer and restart evidence |
| T — Triage | Was there an attempt, successful execution, persistence, or data access? | Correlated application, database, host, identity, and network evidence |
| C — Contain | What must be isolated, revoked, rotated, or preserved? | Approved containment actions and chain of custody |
| H — Harden | Which controls reduce repeat exploitation and blast radius? | Restricted exposure and egress, segmentation, least privilege, monitoring |
| Proof | Is the known path closed, the application functional, and the environment monitored? | Safe retest, functional checks, alert evidence, accepted residual risk |
The framework prevents two common failures: declaring victory after a patch status turns green, and declaring compromise from one weak indicator. It requires a separate answer for remediation, historical exposure, recovery, and proof.
Start by defining an authorized penetration-testing scope with named systems, permitted techniques, data-handling rules, maintenance windows, stop conditions, and Oracle-specific operational constraints. Do not run a leaked or public exploit against production simply to obtain a screenshot.
Validation should combine evidence rather than depend on one scanner result:
For environments that change frequently, continuous penetration testing can revalidate authorized attack paths after material changes. It complements patch and monitoring evidence; it does not replace historical compromise assessment or DFIR.
Escalate when the organization lacks EBS-specific patching experience, historical logs are incomplete, evidence points to code execution or data access, volatile-memory analysis is needed, business interruption constrains containment, or legal and contractual decisions depend on defensible scope.
Evaluate penetration testing services for authorized remediation validation and attack-path testing. Use a qualified incident-response or forensic team when the question is whether compromise occurred, how far it spread, or what data was accessed. One provider may offer both capabilities, but the statement of work and evidence standards should keep the roles distinct.
Oracle lists supported EBS versions 12.2.3 through 12.2.14 as affected. Earlier unsupported releases were not tested through the alert program; Oracle says earlier versions are likely affected and recommends upgrading to a supported release.
Yes. CISA added it to the Known Exploited Vulnerabilities Catalog on October 6, 2025 and marks known ransomware-campaign use as “Known.” Public reporting also documents exploitation against Oracle EBS environments before the emergency patch was available.
No. Oracle lists HTTP as the affected protocol and explains that a listed protocol includes its secure variants unless the secure variant is the only one affected. TLS protects data in transit; it does not correct the vulnerable server-side behavior.
It is the prerequisite stated in Oracle’s original CVE-2025-61882 emergency alert. In 2026, administrators should check the current EBS Patch Availability Document and My Oracle Support for the correct cumulative path, supersedence, and platform-specific requirements rather than blindly replaying the original one-off sequence.
No. Patching reduces future exploitability but does not determine whether exploitation occurred earlier. Review historical application, database, host, identity, and network evidence; scope possible data access; contain confirmed activity; and validate the repaired environment safely.
No. The extortion activity claimed affiliation with the CL0P brand and had overlaps with activity associated with FIN11, but Google Threat Intelligence Group did not formally attribute all activity to a tracked group. Describe it as CL0P-branded activity unless incident-specific evidence supports a narrower conclusion.
CVE-2025-61882 remains operationally relevant because an affected system could have been exposed before the patch, because EBS environments can contain high-value business data, and because a clean installation record does not answer the compromise question. The defensible outcome is a complete chain: current Oracle-supported patching, evidence-led historical review, proportionate containment, hardening, and safe validation.
If you need help defining an authorized Oracle EBS validation scope or testing whether remediated attack paths are actually closed, DeepStrike can support a controlled engagement. If compromise is suspected, begin with incident-response and forensic requirements before treating the work as a routine pentest.
*This article provides defensive technical guidance. Confirm patch applicability and installation instructions with Oracle Support and My Oracle Support. Perform security testing only with explicit authorization and approved rules of engagement.*
Mohammed Khalil is a Cybersecurity Architect at DeepStrike, specializing in advanced penetration testing and offensive security operations. With certifications including CISSP, OSCP, and OSWE, he has led numerous red team engagements for Fortune 500 companies, focusing on cloud security, application vulnerabilities, and adversary emulation. His work involves dissecting complex attack chains and developing resilient defense strategies for clients in the finance, healthcare, and technology sectors.

Stay secure with DeepStrike penetration testing services. Reach out for a quote or customized technical proposal today
Contact Us