A remotely-exploitable critical vulnerability in the Cisco SD-WAN vManage software could allow unauthenticated attackers to retrieve information from vulnerable instances.
Tracked as CVE-2023-20214 (CVSS score of 9.1), the vulnerability exists because the REST API feature of vManage does not sufficiently validate requests.
The vManage API allows administrators to configure, control, and monitor Cisco devices over the network.
An attacker could trigger the vulnerability by sending a crafted API request to a vulnerable instance, to retrieve information from vManage, or send information to it.
“A vulnerability in the request authentication validation for the REST API of Cisco SD-WAN vManage software could allow an unauthenticated, remote attacker to gain read permissions or limited write permissions to the configuration of an affected Cisco SD-WAN vManage instance,” Cisco explains in an advisory.
The web-based management interface and the CLI are not impacted by this security defect, the tech giant explains.
To hunt for attempts to access the REST API, administrators are advised to examine a log file. The existence of requests in the log, however, does not indicate unauthorized access, Cisco explains.
The tech giant notes that, while there are no workarounds to address this bug, implementing access control lists (ACLs) to limit vManage access mitigates the issue.
“In cloud hosted deployments, access to vManage is limited by ACLs that contain permitted IP addresses. Network administrators should review and edit the permitted IP addresses in the ACLs. In on-premises deployments, vManage access can be limited in a similar way by using ACLs and configuring permitted IP addresses,” Cisco explains.
The vulnerability has been addressed with the release of SD-WAN vManage versions 188.8.131.52, 184.108.40.206, 220.127.116.11, 18.104.22.168, 22.214.171.124, and 126.96.36.199. Versions 18.3 to 188.8.131.52 are not affected. Customers using SD-WAN vManage versions 20.7 and 20.8 are advised to migrate to a patched version.
Cisco’s security team says it is not aware of this vulnerability being exploited in attacks.