Enterprise Cybersecurity IntelligenceMonday

An enterprise cybersecurity intelligence company.For security and technology leaders.

Security.io Intelligence

What changed, why it matters,
and how it evolved.

Network Security · Executive briefing

Check Point gateway exploit chain reaches root, persistence and an ESXi wiper

CISA's Friday KEV addition created a Monday remediation and forensic-triage deadline for a Check Point root-escalation flaw used after PHP2Shell access in a gateway-to-ESXi destructive chain.

Vulnerability ManagementNetwork SecurityIncident Response
Why it is in today’s brief

The gateway attacks began in late October, but Friday's KEV addition created a new Monday deadline and made forensic triage part of the required response. The decision changed from following a vendor investigation to proving that patched gateways were not already persistent footholds into ESXi. It warrants inclusion because the chain combines perimeter access, root privileges and demonstrated destructive impact without exceeding the edition's vulnerability limit.

Read first

Patch CVE-2026-123039 through Check Point's product-specific guidance, but close the issue only after gateway, firewall-log and ESXi hunting rules out persistence and destructive follow-on activity.

Act now

Inventory internet-facing Check Point Mobile Access portals.

Accountable owner

CISO with the network-security, firewall, virtualisation and incident-response owners

Decision horizon

Immediate: meet the remediation and forensic-triage requirement today; escalate positive indicators without waiting for patch completion.

AssessmentHigh confidence
Emerging riskConfirmation of destructive activity outside the first wave, additional affected products, revised hotfix guidance or higher-confidence attribution would change scoping and recovery priorities.

What happened

On October 30, 2026, Check Point observed the first exploitation wave against Israeli organisations; a broader second wave followed on November 2, 2026. The chain used PHP2Shell vulnerabilities CVE-2026-109618 and CVE-2026-109619 for initial access, then CVE-2026-123039 to escalate through DynamicCLI to root. The second wave reached multiple countries and sectors, but Check Point reported no evidence of destructive activity in that wave.

On November 12, 2026, Check Point published its technical investigation of the gateway-to-ESXi attack chain. Check Point found upload_progress_pwn in session files under /opt/CPcvpn-R81.20/tmp/ and webshells matching zz*.tpl with the hardcoded key zz2026. The persistence chain used /tmp/CPconfigdaemon, slot_CPconfigdaemon and /lib/libcpfwcommon.so before SSH movement to VMware ESXi. The destructive payload was destroy_v6, and the note was DREAM_ANNEXATION.txt.

On November 13, 2026, CISA added CVE-2026-123039 to the Known Exploited Vulnerabilities Catalog with a federal due date of November 16, 2026. Check Point published 130.94.19[.]204 and 130.94.19[.]168 as infrastructure indicators. The SHA-256 for slot_CPconfigdaemon is 4171b521fa6441382aa2fa014a08e18ff9a6e0ab6096c9144072bf46b37a2ec8, and the destroy_v6 IOC hash is c7d9477309fac6c638a1128442056a242de086996add8cb49cd6496c36003237. Attribution posture: Check Point says the tooling overlaps with PhiliKit, previously associated with UNC5221, but does not assess the campaign’s true operator with high confidence.

Why this matters now

A security gateway sits at a privileged network boundary and often has reachability into management networks. In the investigated intrusion, attackers moved from an exposed portal to root on the gateway, established persistent access, scanned internally and reached VMware ESXi. That sequence turns a perimeter patch decision into a potential infrastructure-recovery decision.

CISA’s addition matters because it confirms active exploitation and requires federal agencies to conduct forensic triage rather than merely install a fix. Enterprises outside the directive should adopt the same separation: a patched gateway can still contain webshells, a persistent implant or evidence of movement into the virtualisation estate.

The destructive activity was observed in only part of the documented campaign, and Check Point did not attribute the operator with high confidence. That uncertainty should constrain claims about motive, but it should not reduce urgency where the published gateway or ESXi indicators are present.

The decision for security leaders

Require separate evidence for remediation and compromise status. The change record proves that the hotfix was applied; gateway files, processes, session artefacts, firewall logs and ESXi access records determine whether the system was already used.

Prioritise gateways exposing Mobile Access, then inspect internal reachability from those devices. Unauthorised SSH from a gateway to ESXi, unexplained firewall-log gaps or the published persistence artefacts should immediately move the case into incident response.

Prepare for destructive recovery if the wiper or ESXi movement is found. Isolate the gateway and affected hypervisor management paths, preserve evidence, validate clean recovery copies and engage executive continuity owners before restoration.

Evidence of closure

  • Asset inventory identifies every gateway exposing Mobile Access during the campaign windows.
  • Change evidence confirms the applicable Check Point hotfix on each affected product.
  • Gateway hunt report records zero indicators or an incident case for each match.
  • ESXi access review shows no unauthorised gateway-originated SSH or destructive artefacts.

The Security.io assessment

CISA and Check Point establish active exploitation of CVE-2026-123039. Check Point additionally documents successful compromise and destructive activity in investigated first-wave cases, but the sources do not establish that every targeted gateway was breached.

The cited research and CISA alert do not publish a single universal fixed build; Check Point directs customers to product-specific November Jumbo Hotfix guidance. Asset owners must therefore validate the exact product and software train against the support advisory rather than rely on a generic version threshold.

The attribution evidence is deliberately inconclusive. Tooling and infrastructure overlaps can guide hunting, but the politically themed note and destructive behaviour do not justify a confident actor or motive claim. Operational decisions should rest on the exploit chain and local evidence.

Questions for the morning meeting

  • Which internet-facing gateways exposed Mobile Access during the campaign windows?
  • Can firewall administrators prove both hotfix deployment and compromise triage?
  • Which ESXi hosts accept SSH connections or service-account access from gateways?
  • Are isolated recovery copies available for workloads behind affected gateways?

Related intelligence

Shared decision context