Cisco Security Agent

The Management Console (webagent.exe) in Cisco Security Agent 5.1, 5.2, and 6.0 before 6.0.2.145 allows remote attackers to create arbitrary files and execute arbitrary code via unspecified parameters in a crafted st_upload request.

Buffer overflow in a certain driver in Cisco Security Agent 4.5.1 before 4.5.1.672, 5.0 before 5.0.0.225, 5.1 before 5.1.0.106, and 5.2 before 5.2.0.238 on Windows allows remote attackers to execute arbitrary code via a crafted SMB packet in a TCP session on port (1) 139 or (2) 445.

Cisco Security Agent (CSA) for Linux 4.5 before 4.5.1.657 and 5.0 before 5.0.0.193, as used by Unified CallManager (CUCM) and Unified Presence Server (CUPS), allows remote attackers to cause a denial of service (resource consumption) via a port scan with certain options.

Unspecified vulnerability in Cisco Security Agent 5.2 before 5.2.0.285, when running on Linux, allows remote attackers to cause a denial of service (kernel panic) via "a series of TCP packets."

Unspecified vulnerability in Cisco Security Agent (CSA) 4.5.0 and 4.5.1 agents, when running on Windows systems, allows local users to bypass protections and gain system privileges by executing certain local software.

The (1) TTLS CHAP, (2) TTLS MSCHAP, (3) TTLS MSCHAPv2, (4) TTLS PAP, (5) MD5, (6) GTC, (7) LEAP, (8) PEAP MSCHAPv2, (9) PEAP GTC, and (10) FAST authentication methods in Cisco Secure Services Client (CSSC) 4.x, Trust Agent 1.x and 2.x, Cisco Security Agent (CSA) 5.0 and 5.1 (when a vulnerable Trust Agent has been deployed), and the Meetinghouse AEGIS SecureConnect Client store transmitted authentication credentials in plaintext log files, which allows local users to obtain sensitive information by reading these files, aka CSCsg34423.

Cisco Secure Services Client (CSSC) 4.x, Trust Agent 1.x and 2.x, Cisco Security Agent (CSA) 5.0 and 5.1 (when a vulnerable Trust Agent has been deployed), and the Meetinghouse AEGIS SecureConnect Client do not properly parse commands, which allows local users to gain privileges via unspecified vectors, aka CSCsh30624.

Directory traversal vulnerability in the Management Center for Cisco Security Agents 6.0 allows remote authenticated users to read arbitrary files via unspecified vectors.

Cisco Secure Services Client (CSSC) 4.x, Trust Agent 1.x and 2.x, Cisco Security Agent (CSA) 5.0 and 5.1 (when a vulnerable Trust Agent has been deployed), and the Meetinghouse AEGIS SecureConnect Client do not drop privileges when the help facility in the supplicant GUI is invoked, which allows local users to gain privileges, aka CSCsf14120.

Cisco Secure Services Client (CSSC) 4.x, Trust Agent 1.x and 2.x, Cisco Security Agent (CSA) 5.0 and 5.1 (when a vulnerable Trust Agent has been deployed), and the Meetinghouse AEGIS SecureConnect Client allows local users to gain SYSTEM privileges via unspecified vectors in the supplicant, aka CSCsf15836.

Cisco Secure Services Client (CSSC) 4.x, Trust Agent 1.x and 2.x, Cisco Security Agent (CSA) 5.0 and 5.1 (when a vulnerable Trust Agent has been deployed), and the Meetinghouse AEGIS SecureConnect Client use an insecure default Discretionary Access Control Lists (DACL) for the connection client GUI, which allows local users to gain privileges by injecting "a thread under ConnectionClient.exe," aka CSCsg20558.

SQL injection vulnerability in the Management Center for Cisco Security Agents 5.1 before 5.1.0.117, 5.2 before 5.2.0.296, and 6.0 before 6.0.1.132 allows remote authenticated users to execute arbitrary SQL commands via unspecified vectors.

The buffer overflow trigger in Cisco Security Agent (CSA) before 4.0.3 build 728 waits five minutes for a user response before terminating the process, which could allow remote attackers to bypass the buffer overflow protection by sending additional buffer overflow attacks within the five minute timeout period.

Cisco Security Agent (CSA) 4.5 allows remote attackers to cause a denial of service (system crash) via a crafted IP packet.