Environments

An environment represents one PostgreSQL database that Drift Scanner watches on your behalf. You can register many environments — one per database you want to track drift on, typically one per deploy target (production, staging, analytics replica, and so on).

JDBC URL format

The URL must begin with jdbc:postgresql:// and include a host, port, and database name:

jdbc:postgresql://<host>:<port>/<database>[?param=value&...]

Examples:

jdbc:postgresql://prod-db.example.com:5432/myapp
jdbc:postgresql://db.internal.example.net:5432/app?sslmode=require

The URL is validated at registration and again on every scan.

Allowed hosts

For security, the host portion cannot resolve to any of the following:

• Loopback (127.0.0.0/8, ::1)
• Private IPv4 (10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16)
• Link-local (169.254.0.0/16 — also includes AWS, GCP, and Azure metadata endpoints)
• Carrier-grade NAT (100.64.0.0/10)
• IPv6 link-local (fe80::/10) and unique-local (fc00::/7)
• Multicast and unspecified addresses (0.0.0.0, ::)

If your database is on a private network, expose it to Drift Scanner through a public endpoint (with TLS and IP allowlisting) or wait for the bring-your-own-agent mode on the roadmap.

Database user permissions

Drift Scanner only reads from pg_catalog. A read-only role with the minimum permissions below is enough:

CREATE USER arcnull_scanner WITH PASSWORD '<strong-password>';
GRANT CONNECT ON DATABASE myapp TO arcnull_scanner;
GRANT USAGE ON SCHEMA pg_catalog TO arcnull_scanner;

We do not need DDL rights and never issue schema modifications. Your credentials are encrypted at rest using AES-256-GCM before persistence.

What gets captured

Each snapshot includes:

• Tables — name, schema, kind
• Columns — name, type, default, nullability, ordinal position
• Indexes — name, columns, unique flag, method
• Constraints — primary keys, foreign keys, unique, check
• Sequences — name, start, increment
• Views — name, definition
• Functions — name, argument types, return type, language

Snapshots include counts of tables, columns, and indexes at the top level for quick at-a-glance reporting.

What is not captured

We intentionally ignore:

• Table data. We only read metadata from pg_catalog.
• User and role grants. On the roadmap.
• Extensions. On the roadmap.

Tier caps

If you hit the cap, the API returns a 400 with a plan-limit message. Upgrade from the dashboard or deactivate an environment you no longer need.

Cron schedule

Each environment has its own cron expression. The default is hourly (0 0 * * * *). You can set anything that matches Spring's cron format — for example, 0 */15 * * * * for every fifteen minutes. Tier-based rate limits still apply.

Deactivating an environment

Deactivating keeps history intact but stops future scans. It does not delete the environment. You can reactivate later from the dashboard.