A PostgreSQL integration connects a Postgres database to your tenant so it can be queried as a catalog on your Trino query engines. Register the connection details once, then apply it to a query engine to start using it.

Before you start

  • Your data plane must be able to reach the database over TCP — a public endpoint, a peered network, or an SSH tunnel to a private one.
  • A Postgres user Dippr can log in as. It needs at least SELECT on the schemas you want to expose (schema discovery reads information_schema.columns, excluding pg_catalog and information_schema).

Register the integration

From Integrations, choose Register integration and select PostgreSQL.
verify-full does not survive an SSH tunnel: the driver checks the certificate against the tunnel’s own address, which no real database certificate names, so the connection fails at connect time. Use require instead for a tunnelled source — you keep end-to-end encryption, and the tunnel’s own host-key verification already confirms which machine you’re talking to.
Capabilities available for this kind: driver (required), federation, and MCP. Federation is what makes the database a catalog on your query engines. MCP is recorded on the integration but not yet wired to any tool: the deployed MCP server ships no per-source tool providers in v1, so selecting it does not expose this Postgres source to agents directly today.
The Register integration dialog with kind PostgreSQL selected

The Register integration dialog with kind PostgreSQL selected.

Test connection

Runs SELECT 1 followed by SELECT version() against the stored (or, before saving, the drafted) credentials and reports the server’s version string on success. A failure reports the driver’s own error — a wrong password, an unreachable host, or a database that doesn’t exist.

In Explore

Once a query engine applies this integration, it appears as a catalog named after the integration’s display name (lowercased, non-alphanumeric characters collapsed to underscores; a name that collides with another catalog gets a short suffix; see How catalogs are named). Every schema and table your Postgres user can read from information_schema appears under it, with column names, types and nullability.

Limitations

  • Discovery reports schemas and tables only — no row counts, indexes or constraints.
  • What you can query or change through the engine is governed by your own roles and grants, not by this record.

Equivalent dip command

Register a Postgres source

Troubleshooting

SSH tunnels

Reach a database sitting behind a bastion, in a private network.

Test connection and secrets

How Test connection works, and how your credentials are stored.