Grandstream UCM6204 any version

Grandstream UCM6200 series firmware version 1.0.20.23 and below is vulnerable to OS command injection via HTTP. An authenticated remote attacker can bypass command injection mitigations and execute commands as the root user by sending a crafted HTTP POST to the UCM's "New" HTTPS API.

Grandstream UCM6200 series firmware version 1.0.20.23 and below is vulnerable to OS command injection via SSH. An authenticated remote attacker can execute commands as the root user by issuing a specially crafted "unset" command.

The UCM6200 series 1.0.20.22 and below stores unencrypted user passwords in an SQLite database. This could allow an attacker to retrieve all passwords and possibly gain elevated privileges.

Grandstream UCM6204 before 1.0.19.20 devices allow remote authenticated users to execute arbitrary code via shell metacharacters in the backupUCMConfig file-backup parameter to the /cgi? URI.

Grandstream UCM6200 series firmware version 1.0.20.23 and below is vulnerable to OS command injection via HTTP. An authenticated remote attacker can execute commands as the root user by sending a crafted HTTP GET to the UCM's "Old" HTTPS API.

Grandstream UCM6204 before 1.0.19.20 devices allow remote authenticated users to conduct SQL injection attacks via the sord parameter in a listCodeblueGroup API call to the /cgi? URI.

The Grandstream UCM6200 series before 1.0.20.22 is vulnerable to an SQL injection via the HTTP server's websockify endpoint. A remote unauthenticated attacker can invoke the challenge action with a crafted username and discover user passwords.

The Grandstream UCM6200 series before 1.0.20.22 is vulnerable to an SQL injection via the CTI server on port 8888. A remote unauthenticated attacker can invoke the challenge action with a crafted username and discover user passwords.

The Grandstream UCM6200 series before 1.0.20.22 is vulnerable to an SQL injection via the HTTP server's websockify endpoint. A remote unauthenticated attacker can invoke the login action with a crafted username and, through the use of timing attacks, can discover user passwords.