Cisco Expressway Flaw Exploited After Public PoC Appeared
Attackers used a critical vulnerability in Cisco Expressway to gain access to a Russian organization shortly after proof-of-concept exploit code became publicly available, cybersecurity firm F6 reports. The activity involved internet-facing Expressway gateways and subsequent movement to other nodes inside the organization’s network.
The flaw, tracked as CVE-2026-76466, carries a CVSS score of 9.8. It can allow a remote attacker with network access to an affected device to execute commands as root without authentication. F6 says it investigated the attack, but did not identify the victim or link the activity to a known threat actor.
Clustered Expressway systems are affected
Expressway gateways support remote access to enterprise collaboration services, including video and voice calls and instant messaging. The vulnerability affects virtual and appliance deployments of both Expressway Core and Expressway Edge, but only when the devices are configured in a cluster.
Cisco disclosed the issue on October 21 and attributed it to the database service feature and its handling of persistent cookies. An attacker who obtains the relevant cookies can access the database service and issue certain remote procedure call (RPC) commands. That access can lead to root-level command execution on the device’s underlying operating system.
Cisco’s stated fix is Expressway X15.5.20. The company says there are no workarounds, and customers running X14 or earlier releases must move to a fixed release. Cisco’s advisory has not been revised since its initial publication and continues to state that the company is unaware of public announcements or malicious use of the vulnerability. That statement predates the exploitation described by F6 and the subsequent public availability of proof-of-concept code on GitHub.
Attackers moved from gateways to internal nodes
According to F6, the attackers initially compromised two internet-exposed Expressway devices running X15.5.3. From one of those systems, they reached two additional cluster nodes located on the organization’s internal network. The report says the attackers obtained root privileges and then prepared tools intended to support lateral movement and the search for sensitive information.
Those tools included custom Python scripts designed to scan the network, find SMB resources, interact with SQL Server and WinRM, and search GitLab repositories. F6 says the apparent objective included obtaining credentials and other sensitive information for exfiltration. The firm also observed that the attackers changed the scripts during the operation, rather than relying only on a fixed set of tools prepared in advance.
F6 noted comments and descriptions of fixes in the scripts’ source code that it considered characteristic of AI-assistant use. That observation concerns the apparent development of the scripts; it does not identify the operators or establish their identity. The firm did not provide a threat-actor attribution.
Patch first; restrict access if deployment must wait
Organizations should identify Expressway Core and Edge systems, confirm whether they are clustered, and prioritize upgrading affected installations to X15.5.20. Cisco identifies no workaround, so firewall restrictions should be treated as a temporary risk-reduction measure rather than a replacement for the vendor’s fixed release.
For organizations unable to patch immediately, F6 recommends blocking connections to TCP ports 4371–4372 at the firewall, allowing access only from IP addresses belonging to trusted cluster nodes. Such restrictions can reduce exposure to those services, but administrators should validate the necessary cluster communications and apply the change in accordance with their environment’s requirements.
F6 points defenders to the Erlang system log on Expressway devices when checking for signs of exploitation. The firm has also published IP addresses and file paths associated with the activity it investigated. Security teams can use those details to guide log review and endpoint or network investigations, while recognizing that indicators from one incident may not capture all activity exploiting the flaw.
The incident illustrates the short interval that can separate public exploit availability from observed attacks against exposed enterprise infrastructure. Operators of clustered Expressway systems should verify their software versions, review access to the affected services, and investigate relevant logs—particularly if systems remain on X15.5.3 or earlier releases.