Amazon Redshift Python Connector

2.1.12 · active · verified Sat Mar 28

redshift-connector is the official Amazon Redshift connector for Python, implementing the Python Database API Specification 2.0. It provides easy integration with data science libraries like pandas and numpy, and supports Redshift-specific features such as IAM and Identity Provider (IdP) authentication. The library is actively maintained with frequent minor releases, currently at version 2.1.12.

Warnings

breaking A security vulnerability (CVE-2025-5279) existed in versions 2.0.872 through 2.1.6, where the `BrowserAzureOAuth2CredentialsProvider` plugin skipped SSL certificate validation for the Identity Provider (IdP). This could allow an actor to intercept token exchange.
Fix: Upgrade to `redshift-connector` version 2.1.7 or later to address this issue.
gotcha Autocommit is `OFF` by default, adhering to the Python DB-API specification. Transactions must be explicitly committed for changes to persist.
Fix: Set `conn.autocommit = True` after connecting, or explicitly call `conn.commit()` after executing statements that modify data.
gotcha Network connectivity issues, including timeouts, are common, especially when connecting from outside an Amazon EC2 instance or running long-duration queries. This can be due to firewall rules, VPC configurations, or client-side TCP/IP timeout settings.
Fix: Ensure Redshift cluster security groups and network ACLs allow inbound/outbound traffic on the Redshift port (default 5439). Increase client system TCP/IP timeout values for long-running queries or idle connections.
gotcha For loading large amounts of data into Redshift, using individual `INSERT` statements can be prohibitively slow and expensive.
Fix: Strongly prefer using the `COPY` command with Amazon S3 staging for bulk data loads. The `redshift-connector` supports this method.
deprecated Amazon Redshift is ending support for creating new Python User-Defined Functions (UDFs) starting with Patch 198. Existing Python UDFs will continue to function but will be fully unsupported after June 30, 2026.
Fix: Migrate existing Python UDFs to alternative solutions before June 30, 2026. For new functionality, consider SQL UDFs or other data processing patterns.

Install

pip install redshift-connector Basic Installation
pip install redshift-connector[full] With Data Science Integrations (pandas, numpy)

Imports

redshift_connector
```
import redshift_connector
```
The common module name to import is `redshift_connector`.

Quickstart

This quickstart demonstrates how to connect to an Amazon Redshift cluster, create a cursor, execute a simple SQL query, fetch the results, and then close the connection. It's recommended to use environment variables for sensitive credentials.

import redshift_connector
import os

try:
    # Connect to Redshift cluster using environment variables for credentials
    conn = redshift_connector.connect(
        host=os.environ.get('REDSHIFT_HOST', 'your-redshift-cluster.example.com'),
        database=os.environ.get('REDSHIFT_DB', 'dev'),
        port=int(os.environ.get('REDSHIFT_PORT', '5439')),
        user=os.environ.get('REDSHIFT_USER', 'awsuser'),
        password=os.environ.get('REDSHIFT_PASSWORD', 'your_password_here')
    )

    # Create a Cursor object
    cursor = conn.cursor()

    # Execute a query
    cursor.execute("SELECT 1 as id, 'Hello, Redshift!' as message;")

    # Retrieve the query result set
    result = cursor.fetchall()
    print("Query Result:", result)

    # Close the cursor and connection
    cursor.close()
    conn.close()

except Exception as e:
    print(f"An error occurred: {e}")

view raw JSON →