Sanitization

Overview

Sanitization runs a SQL transformation on a snapshot to remove or replace sensitive data. It operates as a 3-step pipeline:

Restore the source snapshot to a dedicated sanitization database
Execute the sanitization SQL query against that database
Dump the sanitized database to a new snapshot

The result is a new snapshot with type sanitized that is safe to share with developers, use in CI/CD, or load into non-production environments.

Prerequisites

The target must have two additional fields configured:

Field	Description
`sanitize_dst_url`	Connection URL for the dedicated sanitization database
`sanitize_query_file`	Path to a SQL file containing the sanitization queries

Setup

When adding a target, include the sanitization fields:

clonit targets add \
  --name mydb \
  --src-url "postgres://user:pass@prod:5432/mydb" \
  --dst-url "postgres://user:pass@localhost:5432/mydb_dev" \
  --sanitize-dst-url "postgres://user:pass@localhost:5432/mydb_sanitize" \
  --sanitize-query-file /path/to/sanitize.sql

Writing Sanitization Queries

The sanitization query file is a plain SQL file that runs against the restored snapshot. You can use any valid SQL statements – UPDATE, DELETE, TRUNCATE, etc.

Example `sanitize.sql`

-- Replace email addresses with deterministic fake values
UPDATE users SET email = concat('user', id, '@example.com');

-- Replace password hashes with a known dummy value
UPDATE users SET password_hash = '$2a$10$fake_hash';

-- Remove audit logs entirely
TRUNCATE TABLE audit_logs;

Tips for writing sanitization queries

Use deterministic replacements (e.g., based on id) so the sanitized data is consistent across runs.
Truncate large tables that contain only operational data (logs, events, sessions) to reduce snapshot size.
Test your queries against a copy of the database before using them in production workflows.
Order matters – queries execute sequentially in the order they appear in the file.

Running Sanitization

Sanitize using the latest snapshot

clonit sanitize mydb

Sanitize a specific snapshot by index

clonit sanitize mydb 0

Use clonit snapshots mydb to see available snapshot indices.

Output

After sanitization completes, a new snapshot is created with type sanitized. You can see it in the snapshots list:

clonit snapshots mydb

Example output:

Index  Name                                Type        Size      Created
0      mydb-20260208T120000                snapshot    245 MB    2026-02-08 12:00:00
1      mydb-20260208T120000-sanitized      sanitized   198 MB    2026-02-08 12:05:00

The sanitized snapshot can then be used like any other snapshot – load it to a destination database, push it to cloud storage, or share it with your team.

Workflow Example

A typical sanitization workflow looks like this:

# 1. Build a snapshot from production
clonit build mydb

# 2. Sanitize the snapshot
clonit sanitize mydb

# 3. Load the sanitized snapshot to a dev database
clonit load mydb 1

# 4. Optionally push the sanitized snapshot to cloud storage
clonit push mydb 1