> For the complete documentation index, see [llms.txt](https://v2.dataos.info/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://v2.dataos.info/build/productize/connect-engine/trino/external-trino.md).

# External Trino

Connect Vulcan to your own Trino-compatible endpoint, such as Starburst or self-hosted Trino.

Use this path when you bring your own Trino cluster: Starburst, self-hosted Trino, or any Trino-compatible endpoint running outside DataOS. Vulcan only connects to it; it doesn't provision or manage the cluster. For the DataOS-managed shared cluster, use [Minerva](https://v2.dataos.info/build/productize/connect-engine/trino/minerva.md) instead. To ship a cluster with the product, use [Dedicated Trino](https://v2.dataos.info/build/productize/connect-engine/trino/dedicated-trino.md).

## Core settings

| Setting         | Value              |
| --------------- | ------------------ |
| Connection type | `trino`            |
| Dialect         | `trino`            |
| VDE             | `false` (required) |
| Scheduler       | Local or built-in  |

## Before you start

You need a running Trino endpoint and its host, port, user, catalog, and the authentication method it expects (for example, basic auth over HTTPS). Store the password in an environment variable locally, or a DataOS Secret when deploying.

## Permissions

The Trino user needs `USE` on the source and target catalogs, `CREATE SCHEMA` on the target catalog if Vulcan creates the schema, `SELECT` on source tables and views, and `CREATE TABLE`, `CREATE VIEW`, `SELECT`, `INSERT`, `DELETE`, `DROP TABLE`, `DROP VIEW`, and `ALTER TABLE` on the target schema. `UPDATE` isn't needed: Vulcan uses `DELETE` + `INSERT` for incremental merges. Access control itself is connector/plugin-dependent (none, file-based, OPA, Ranger, or Starburst's own system). Check with whoever operates the cluster which model applies and how grants are expressed there.

## Connection options

Beyond `host`, `port`, `user`, `catalog`, and `password`, the adapter also accepts:

| Option                    | Description                                                                                                             |
| ------------------------- | ----------------------------------------------------------------------------------------------------------------------- |
| `method`                  | Auth method: `basic`, `ldap`, `jwt`, `kerberos`, `certificate`, `oauth`, or `no-auth`.                                  |
| `roles`                   | Catalog-to-role mapping for role-based access control.                                                                  |
| `http_headers`            | Extra HTTP headers sent with each request.                                                                              |
| `session_properties`      | Trino session properties.                                                                                               |
| `retries`                 | Number of request retries.                                                                                              |
| `timezone`                | Connection timezone.                                                                                                    |
| `schema_location_mapping` | Regex-to-location mapping for schema creation, for catalogs that can't infer table locations.                           |
| `catalog_type_overrides`  | Explicit connector type per catalog (for example `iceberg`, `hive`, `delta_lake`), when Vulcan can't reliably infer it. |

## config.yaml

```yaml
vde: false
gateways:
  default:
    connection:
      type: trino
      host: "<trino-host>"
      port: <port>
      user: "{{ env_var('TRINO_USER') }}"
      catalog: "<catalog>"
      http_scheme: https
      method: basic
      password: "{{ env_var('TRINO_PASSWORD') }}"
      verify: true
default_gateway: default
model_defaults:
  dialect: trino
  start: "2024-01-01"
ignore_patterns:
  - "*-deploy.yaml"
```

If your endpoint authenticates differently, set `method` to whichever scheme it enables (`basic`, `ldap`, `jwt`, `kerberos`, `certificate`, `oauth`, or `no-auth`) plus the fields your provider documents. For Starburst and self-hosted specifics, see the [Trino engine manual](https://v2.dataos.info/references/engine-guide/trino-1/external-trino) in References.

## Supported model kinds

`VIEW`, `FULL`, `SEED`, `INCREMENTAL_BY_TIME_RANGE`, and `INCREMENTAL_BY_PARTITION`. For how each kind materializes, see the [Trino engine manual](https://v2.dataos.info/references/engine-guide/trino-1/external-trino).

## Starburst Galaxy specifics

Starburst Galaxy is a managed, Trino-compatible query service — Vulcan connects to an existing cluster and never provisions Starburst compute. The `deployment:` block in `config.yaml` has **no `deployment.trino` and no `deployment.depots`**: credentials flow entirely through a DataOS `Secret` projected with `deployment.use.secrets` and `deployment.use.env_vars`, and catalogs (source and target) must already be attached to the cluster in the Galaxy console before Vulcan can use them — both the read and write catalogs must be attached to the **same** cluster.

{% hint style="warning" %}
Galaxy usernames take a `<email>/<role>` shape — omitting the `/<role>` suffix produces a confusing `404 User not found` rather than an auth error. Vulcan's live catalog-type detection (Iceberg vs. Hive-style DDL) isn't reliable against Galaxy: set `catalog_type_overrides: { <catalog>: iceberg }` explicitly on the gateway connection for any catalog used with `INCREMENTAL_BY_PARTITION`, or Iceberg rejects the Hive-style `partitioned_by` property Vulcan falls back to. Trino's built-in `tpch` catalog is read-only and useful for hello-world testing, but nothing can be materialized into it.
{% endhint %}

| Symptom                                                                            | Likely cause                                                   | Fix                                                               |
| ---------------------------------------------------------------------------------- | -------------------------------------------------------------- | ----------------------------------------------------------------- |
| `TrinoUserError: Table property 'partitioned_by' not supported for Iceberg tables` | Catalog-type auto-detection didn't resolve to `'iceberg'`      | Set `catalog_type_overrides` explicitly for that catalog.         |
| `error 401: access_denied`                                                         | Bad password, or a double-base64-encoded secret in production  | Store the secret as plain text — DataOS encodes it once on apply. |
| `error 404: User not found`                                                        | Username missing the `/<role>` suffix                          | Use `<email>/<role>`.                                             |
| `TrinoUserError(name=CATALOG_NOT_FOUND, ...)`                                      | Catalog not attached to the cluster `STARBURST_HOST` points at | Attach it in the Galaxy console → Clusters → Catalogs tab.        |

For the full Starburst reference — the complete hello-world walkthrough, production deployment stepper, and failure-mode table — see [Starburst](https://v2.dataos.info/references/engine-guide/trino-1/external-trino/starburst) in the Trino engine manual.

Once the connection verifies, continue to [Assets](https://v2.dataos.info/build/productize/assets.md).


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://v2.dataos.info/build/productize/connect-engine/trino/external-trino.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
