Connect a Database

This guide walks you through connecting a database to Slim.io for PII scanning. Slim.io supports PostgreSQL, MySQL, Snowflake, SQL Server, Oracle, IBM DB2, and Databricks.

Time required: 5–15 minutes

Prerequisites:

• Database admin access to create a read-only user
• Network connectivity between your database and Slim.io (or a BYOC agent)
• Slim.io Customer Dashboard access

Supported Databases

Step 1: Create a Read-Only Database User

Before connecting, create a dedicated read-only user for Slim.io. This follows the principle of least privilege — Slim.io only needs SELECT access to scan for PII.

PostgreSQL

CREATE ROLE slimio_scanner LOGIN PASSWORD 'your-secure-password';
GRANT USAGE ON SCHEMA public TO slimio_scanner;
GRANT SELECT ON ALL TABLES IN SCHEMA public TO slimio_scanner;
ALTER DEFAULT PRIVILEGES IN SCHEMA public
  GRANT SELECT ON TABLES TO slimio_scanner;

MySQL

CREATE USER 'slimio_scanner'@'%' IDENTIFIED BY 'your-secure-password';
GRANT SELECT ON your_database.* TO 'slimio_scanner'@'%';
FLUSH PRIVILEGES;

Snowflake

CREATE ROLE SLIMIO_SCANNER;
CREATE USER SLIMIO_USER PASSWORD='your-secure-password'
  DEFAULT_ROLE=SLIMIO_SCANNER;
GRANT ROLE SLIMIO_SCANNER TO USER SLIMIO_USER;
GRANT USAGE ON WAREHOUSE YOUR_WAREHOUSE TO ROLE SLIMIO_SCANNER;
GRANT USAGE ON DATABASE YOUR_DATABASE TO ROLE SLIMIO_SCANNER;
GRANT USAGE ON ALL SCHEMAS IN DATABASE YOUR_DATABASE TO ROLE SLIMIO_SCANNER;
GRANT SELECT ON ALL TABLES IN DATABASE YOUR_DATABASE TO ROLE SLIMIO_SCANNER;

SQL Server

CREATE LOGIN slimio_scanner WITH PASSWORD = 'your-secure-password';
CREATE USER slimio_scanner FOR LOGIN slimio_scanner;
GRANT SELECT TO slimio_scanner;

Oracle

CREATE USER slimio_scanner IDENTIFIED BY "your-secure-password";
GRANT CREATE SESSION TO slimio_scanner;
GRANT SELECT ANY TABLE TO slimio_scanner;

IBM DB2

GRANT CONNECT ON DATABASE TO USER slimio_scanner;
GRANT SELECT ON TABLE schema_name.* TO USER slimio_scanner;

Databricks

For Databricks, generate a Personal Access Token instead of a database user:

1. Log in to your Databricks workspace.
2. Click your profile icon and select Settings.
3. Go to Developer > Access Tokens.
4. Click Generate New Token, add a description (e.g., “Slim.io Scanner”), and set an expiration.
5. Copy the token immediately — it will not be shown again.

Step 2: Start the Connector Wizard

1. Log in to the Slim.io Customer Dashboard at /dash/.
2. Navigate to Connectors in the sidebar.
3. Click Add Connector, or click the database provider tile in the Available Integrations grid.
4. Select your database provider (PostgreSQL, MySQL, Snowflake, etc.).

Step 3: Enter Connection Details

The wizard displays a configuration form specific to your database type.

PostgreSQL, MySQL, SQL Server, Oracle, DB2

Snowflake

Databricks

Step 4: Test the Connection

1. Click Test Connection in the wizard.
2. Slim.io verifies it can connect, authenticate, and query information_schema (or equivalent).
3. On success, you see a green confirmation with the number of discoverable tables.

If the test fails, check:

• Network access: Can the Slim.io service reach your database host and port?
• Credentials: Are the username and password correct?
• Permissions: Does the user have SELECT access?
• SSL: Does your database require SSL? Toggle the SSL switch accordingly.

Step 5: Create the Connector

1. Click Connect to save the connector.
2. The connector appears in your Connectors list with a Pending status.
3. Status transitions to Active once the initial metadata enumeration completes.

Step 6: Verify and Scan

After the connector shows Active status:

• The connector appears in your Connectors list with the database provider icon
• You can trigger a scan immediately by clicking the Scan button
• Slim.io uses schema-first intelligence — it reads column metadata before scanning content, skipping non-text columns (timestamps, booleans, integers) automatically
• Findings are attributed to specific tables and columns

How Database Scanning Works

Slim.io’s database scanning is designed for enterprise scale:

1. Schema Analysis — Reads information_schema to understand column names and data types
2. Column Filtering — Skips columns that cannot contain PII (timestamps, booleans, auto-increment IDs)
3. Column Name Intelligence — Columns named email, ssn, phone, etc. receive a confidence boost
4. Adaptive Sampling — Sample size scales with table size and risk: high-risk tables get deeper scans
5. Server-Side Cursors — Uses streaming cursors (fetchmany) instead of loading full result sets into memory
6. Column-Level Findings — Every finding is attributed to a specific table and column for precise remediation

Troubleshooting

Next Steps

• Run Your First Scan — Trigger a scan and review PII findings
• Create a Custom Classifier — Define patterns for organization-specific sensitive data
• Database Connector Reference — Deep-dive into schema-first scanning and per-database configuration