PostgreSQL COPY TO/FROM PROGRAM Arbitrary Command Execution

Repository contains a single Python exploit script (CVE-2019-9193.py) and a short README with usage examples. The script uses psycopg2 to authenticate to a remote PostgreSQL instance (defaults: 127.0.0.1:5432, database template1, user/password postgres/postgres), checks the server version via `SELECT version()`, and if a `-c/--command` is provided, achieves OS command execution by creating a randomly named table and invoking `COPY <table> FROM PROGRAM '<command>'`, then selecting and printing the resulting output. It attempts cleanup by dropping the created table, including in an exception path. Overall purpose: authenticated RCE against PostgreSQL instances where `COPY ... FROM PROGRAM` is usable, with command output retrieval and basic operational flow (connect → version check → execute → cleanup).

This repository contains a Python exploit script (CVE-2019-9193.py) and a README.md for CVE-2019-9193, a vulnerability in PostgreSQL versions 9.3 through 11.7 that allows authenticated remote code execution. The exploit connects to a target PostgreSQL server using user-supplied credentials and executes arbitrary system commands via the 'COPY FROM PROGRAM' SQL feature, retrieving the output. The script checks the PostgreSQL version to ensure it is vulnerable, then creates a temporary table, executes the command, fetches the output, and cleans up. The README provides usage instructions and example output. The exploit requires network access to the PostgreSQL server and valid credentials. The main attack vector is network-based, targeting the PostgreSQL service (default port 5432). No hardcoded external URLs or IPs are present beyond the default localhost/127.0.0.1 examples.

This repository contains a single Python exploit script (GenPostgresRCEExploit.py) targeting CVE-2019-9193, an authenticated remote code execution vulnerability in PostgreSQL versions 9.3 through 11.7. The exploit requires valid credentials to connect to the target PostgreSQL server and leverages the 'COPY ... FROM PROGRAM' SQL command to execute arbitrary system commands on the server. The script first checks the PostgreSQL version to ensure it is vulnerable, then creates a temporary table, executes the user-supplied command, retrieves the output, and cleans up the temporary table. The repository also includes a README.md with detailed usage instructions and examples, a requirements.txt specifying the psycopg2 dependency, and an MIT license file. The exploit is a proof-of-concept and does not include advanced payloads or automation for post-exploitation. The main attack vector is network-based, requiring access to the PostgreSQL service (default port 5432) and valid authentication credentials.

This repository contains a single Metasploit module: 'PostgreSQL COPY FROM PROGRAM Command Execution'. The module exploits a feature in PostgreSQL 9.3 and above that allows superusers (or users with the 'pg_execute_server_program' privilege) to execute arbitrary system commands via the SQL 'COPY FROM PROGRAM' statement. The exploit works by creating a temporary table, then issuing a 'COPY FROM PROGRAM' command with attacker-supplied payloads, such as reverse shells or arbitrary commands. The module supports multiple platforms (Linux, Windows, OSX) and can deliver payloads appropriate to each. The exploit requires valid credentials and sufficient privileges on the target PostgreSQL server, which must be accessible over the network. The code is weaponized, as it is part of the Metasploit framework and supports customizable payloads. The only endpoint is the PostgreSQL server itself, typically accessed over TCP (default port 5432).

This repository contains a single Python exploit script (GenPostgresRCEExploit.py) targeting CVE-2019-9193, an authenticated remote code execution vulnerability in PostgreSQL versions 9.3 through 11.7. The exploit requires valid credentials to connect to the target PostgreSQL server and leverages the 'COPY ... FROM PROGRAM' SQL command to execute arbitrary system commands on the server. The script checks the server version for vulnerability, connects using user-supplied parameters (IP, port, database, username, password), and executes a user-specified command, returning its output. The repository also includes a README in both English and Russian, a requirements.txt specifying the psycopg2 library, and an MIT license. The exploit is a proof-of-concept and is not weaponized, but demonstrates the core vulnerability and its impact.

This repository contains a Python proof-of-concept exploit for CVE-2019-9193, targeting PostgreSQL versions 9.3 through 12.3. The exploit leverages the 'COPY ... FROM PROGRAM' SQL command, which allows an authenticated user to execute arbitrary system commands on the database server. The main script, 'cve-2019-9193.py', accepts command-line arguments for the target IP, port, database name, user credentials, and the system command to execute. Upon successful authentication, it checks the PostgreSQL version and, if a command is provided, creates a temporary table, executes the command via the vulnerable SQL feature, retrieves the output, and cleans up. The exploit requires valid database credentials and network access to the PostgreSQL server. The README provides context, references, and usage examples. No hardcoded external endpoints are present; the script is designed to be run by an attacker who specifies the target. The repository is structured simply, with one exploit script and a README.

This repository contains a Python exploit script (portgres_rce.py) targeting CVE-2019-9193, a vulnerability in PostgreSQL that allows code execution via the 'COPY FROM PROGRAM' feature when used by a superuser or a user in the 'pg_execute_server_program' group. The exploit works by connecting to a vulnerable PostgreSQL server and executing a Perl one-liner to establish a reverse shell from the server to the attacker's machine. The attacker is prompted for the target (RHOST, RPORT) and their own (LHOST, LPORT) connection details. The script sets up a listener on the attacker's machine and, upon successful exploitation, provides an interactive shell for arbitrary command execution on the server. The repository is minimal, with only one code file, and is focused solely on this exploit. No detection or fake code is present.