Home › Vulnerabilities

Check Point Multiple Products: Exploited Flaw CVE-2026-123039

Published November 13, 2026 · Last updated November 13, 2026

CVE KEV Patch

What happened: On November 13, 2026, the U.S. Cybersecurity and Infrastructure Security Agency (CISA) added CVE-2026-123039, a flaw in Check Point Multiple Products, to its Known Exploited Vulnerabilities (KEV) catalog. The catalog lists flaws CISA has evidence attackers are using. CISA calls this one the "Check Point Multiple Products Incorrect Permission Assignment for Critical Resource Vulnerability."

CISA's summary: "Check Point Firewall, Multi-Domain Security Management Server, Scalable Platforms, Security Gateway, and Security Management Server contain an incorrect permissions assignment for critical resources vulnerability when the Gaia OS DynamicCLI daemon runs as root and listens on a world-writable local socket without authenticating connecting processes. An authenticated local user can send a command to the socket and have it executed as root."

CVE-2026-123039 is a local privilege escalation flaw in Gaia OS, the operating system that runs Check Point firewalls and security management servers. Check Point says it has "observed an active exploit of this vulnerability against a couple of dozen customers." On November 13, 2026, the U.S. Cybersecurity and Infrastructure Security Agency (CISA) added it to its Known Exploited Vulnerabilities (KEV) catalog, with a due date of November 16, 2026. CISA lists known ransomware campaign use as "Unknown."

The bug does not let an outsider break in by itself. It turns a foothold into full control: anyone who can already run code on the appliance as an ordinary local user can become root, the most powerful account on the system.

What goes wrong

Gaia's extended command line, called Dynamic CLI, is served by a background program named dynamic_server that runs as root. It takes requests through a local socket file, /tmp/dynamic_socket. According to Check Point's advisory (sk1000362), on affected systems that socket is created world-writable, with no check on which process is connecting, and the daemon "executes whatever command string it receives as root." The result, in Check Point's words: "any authenticated local user can run arbitrary commands as root by writing a single request to the socket."

Dynamic CLI exists to add extra commands to Clish, Gaia's regular administrative shell; Check Point documents those commands in a separate article, sk144112. That convenience layer is what holds root privileges here.

CISA names the entry "Check Point Multiple Products Incorrect Permission Assignment for Critical Resource Vulnerability." It and the CVE record classify the weakness as incorrect permission assignment for a critical resource (CWE-732) and missing authorization (CWE-862).

Check Point is clear that this is not a remote entry point. The flaw "requires the attacker to already have local code execution on the appliance." In practice that means it is most dangerous as a second step, after an attacker gets in some other way, for example through a stolen login or a separate bug.

Who is affected

Check Point's advisory lists these as affected:

  • Check Point Force Firewalls
  • Check Point Management and Log Servers
  • Multi-Domain Security Management and Multi-Domain Log Server
  • Maestro Orchestrator

The CVE record uses Check Point's Quantum product names, listing Quantum Security Gateway and Quantum Security Management. Affected versions are R82.20, R82.10, R82, R81.20, R81.10 and R81. Check Point marks R81.10 and R81 as end of support. Spark Firewalls and Smart-1 Cloud are listed as not affected.

The fix

Check Point says the problem is fixed in these Jumbo Hotfix Accumulator releases:

  • R82.20: Take 14 or later
  • R82.10: Take 52 or later
  • R82: Take 132 or later
  • R81.20: Take 174 or later

The advisory lists no Jumbo take for R81.10 or R81, the two end-of-support versions. Owners of those systems should plan an upgrade to a supported version and use the mitigation below in the meantime.

The CVE record, also published by Check Point, describes the affected builds a little differently, for example "R82.10 with Jumbo Hotfix Take 50 or below." Where the two do not line up exactly, use the advisory's minimum fixed take as the target.

According to the advisory, the fix blocks and logs unauthenticated calls. After patching, the command ls -l /tmp/dynamic_socket should show the socket owned by root with mode 0660, not world-writable 0777.

If you cannot patch yet

Check Point's mitigation is to stop the dynamic_server process and keep it from restarting, using dbset process:dynamic_server followed by dbset :save. This breaks the Dynamic CLI commands described in sk144112, so expect some Clish commands to stop working until you patch. Once the fix is installed, the advisory's command to turn the process back on is dbset process:dynamic_server t followed by dbset :save. To see whether the daemon is running, look for it in the process list from Expert mode; the advisory's sample output shows it running as /usr/bin/dynamic_server.

Checking for compromise

Because exploitation has been seen, patching is not the end of the job. The vulnerable daemon logs every command it receives to /var/log/dynamicCli.log. Check Point says legitimate requests arrive wrapped in a standard Clish profile line, and "any command that does not use this wrapper may indicate that a command was executed as root through the socket and requires investigation." That normal wrapper loads a shared Check Point profile script and switches off shell filename expansion before calling a Gaia binary. The advisory gives three searches to run on each Force Firewall or cluster member: one for requests missing that wrapper; one for a #return_value# marker, which points to the daemon's system() call and which Check Point says is "uncommon or absent in legitimate traffic"; and one for words often seen in attacks, such as whoami, /etc/shadow, ncat, /dev/tcp, curl, wget, base64, useradd or authorized_keys. A hit on any of these deserves a closer look at that device, its accounts and its recent configuration changes.

On patched systems, a log line reading "Rejected connection from unauthorized uid" with an unexpected user ID means an escalation attempt was blocked. Check Point's example shows user ID 99, which it identifies as the "nobody" account. The advisory suggests matching bursts of those lines against portal, web and authentication logs to find where the attempt came from.

Severity and who scored it

Check Point, the CVE Numbering Authority for its own products, scored the flaw CVSS 3.1 7.8 (High), vector AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H. NVD shows that same Check Point score as a Secondary source; NVD had not added its own score when we checked, and its analysis was still in progress. The score reflects local access with low privileges and full impact on the device once exploited.

CISA's own decision data for the CVE marks exploitation as "active," automation as "no," and technical impact as "total." In plain terms, CISA considers it actively exploited, not automatable at scale, and able to give complete control of an affected system.

The rating is a starting point, not a ceiling. A firewall or management server sits at the center of a network's defenses, so root on that box can mean control of the rules that protect everything behind it.

What to do

  • Firewall and security administrators: install the Jumbo Hotfix take listed above for your version, then confirm the socket shows mode 0660.
  • If patching must wait: stop dynamic_server as described in the advisory and accept the loss of the affected Dynamic CLI commands for now.
  • Owners of R81.10 or R81: no fixed take is listed. Apply the mitigation and schedule an upgrade to a supported release.
  • Security and incident response teams: review /var/log/dynamicCli.log on every gateway and cluster member using the advisory's checks, and investigate any unexpected commands.
  • U.S. federal agencies: the KEV entry says to apply mitigations according to the vendor's instructions by November 16, 2026, following CISA's BOD 26-04 guidance and its forensics triage requirements, or to stop using the product if no mitigation is available.

Timeline

  • November 8, 2026: Check Point creates advisory sk1000362.
  • November 12, 2026: the CVE record is published and the advisory is last updated.
  • November 13, 2026: CISA adds the flaw to the KEV catalog.

Sources: Check Point advisory sk1000362; the CISA Known Exploited Vulnerabilities catalog; the CVE record published by Check Point; NVD.

What the National Vulnerability Database says

NVD first published CVE-2026-123039 on November 12, 2026. Its description, from the record's source ([email removed]), reads: "The Gaia OS DynamicCLI daemon runs as root and listens on a world-writable local socket (/tmp/dynamic_socket) without authenticating connecting processes. An authenticated local user can send a command to the socket and have it executed as root."

How serious is CVE-2026-123039?

The CVSS 3.1 base score is 7.8 out of 10 (High), scored by checkpoint.com. In plain terms: an attacker can reach the Multiple Products weakness only with local access to the system; the attacker needs a low-privilege account; no one has to click or open anything. The listed impact on Multiple Products data secrecy is high, on data integrity high, and on availability high. The full vector is CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H.

Which Multiple Products versions are affected?

The CVE record lists these affected ranges:

  • version R82.10 with Jumbo Hotfix Take 50 or below
  • version R82.20 with Jumbo Hotfix Take 10 or below
  • version R82 with Jumbo Hotfix Take 130 or below
  • version R81.20 with Jumbo Hotfix Take 173 or below
  • version R82.10 with Jumbo Hotfix Take 50 or below
  • version R82.20 with Jumbo Hotfix Take 10 or below
  • version R82 with Jumbo Hotfix Take 130 or below
  • version R81.20 with Jumbo Hotfix Take 173 or below

If your Multiple Products version falls in one of these ranges, check Check Point's advisory for the fixed release that applies to you.

What kind of weakness is it?

CISA and NVD map CVE-2026-123039 to the following weakness types in MITRE's Common Weakness Enumeration (CWE):

  • CWE-732: Incorrect Permission Assignment for Critical Resource. MITRE's definition: "The product specifies permissions for a security-critical resource in a way that allows that resource to be read or modified by unintended actors."
  • CWE-862: Missing Authorization. MITRE's definition: "The product does not perform an authorization check when an actor attempts to access a resource or perform an action."

CISA required action for CVE-2026-123039

CISA's required action for this KEV catalog entry, quoted from the catalog record, is: "Apply mitigations in accordance with vendor instructions, ensuring compliance with CISA’s BOD 26-04 Prioritizing Security Updates Based on Risk (see URL in Notes) guidance and CISA’s “Forensics Triage Requirements” (see URL in Notes). Follow applicable BOD 26-04 guidance for cloud services or discontinue use of the product if mitigations are unavailable. Stakeholders are responsible for evaluating each asset's internet exposure and ensuring adherence to BOD 26-04 patching guidelines."

Timeline for CVE-2026-123039

  • November 12, 2026: NVD publishes CVE-2026-123039.
  • November 13, 2026: CISA adds CVE-2026-123039 to the KEV catalog.
  • November 13, 2026: latest change to the NVD record (status: Undergoing Analysis).
  • November 16, 2026: due date for U.S. federal civilian agencies to act on CVE-2026-123039.

Advisories and technical references

The NVD record for CVE-2026-123039 links to these references; tags in brackets are NVD's labels:

What to do

  • Run Check Point Multiple Products? Follow Check Point's own advisory (linked under Sources) and apply the fix or mitigation it lists. CISA's required action for this entry is to apply the vendor's mitigations, or stop using the product if no mitigation is available.
  • Federal civilian agencies must act by November 16, 2026, CISA's due date for this entry. Other organizations can use that date as a deadline for Multiple Products too.
  • Ransomware use: CISA lists known ransomware campaign use for CVE-2026-123039 as "Unknown".
  • Don't run Multiple Products yourself? You don't need to install anything on your phone or home computer for this. If a call, text, or pop-up says your device is infected by "CVE-2026-123039" and asks for payment or remote access, that is a scam pattern, not a Check Point alert.

CVE-2026-123039 record details from CISA and NVD

  • CISA forensic triage flag for CVE-2026-123039: Yes
  • Attack complexity for CVE-2026-123039: low
  • CVE-2026-123039 CVSS scope: unchanged
  • NVD sub-scores for CVE-2026-123039: exploitability 1.8, impact 5.9.
  • Who mapped the weakness type: the address listed on checkpoint.com (Secondary)
  • References in the NVD record: 2.
  • CISA SSVC decision points for CVE-2026-123039 (from NVD): exploitation=active; automatable=no; technicalImpact=total (role: CISA Coordinator, SSVC 2.0.3).

Sources

General education only, not security advice. Facts checked against the sources above on the date shown.