Skip to content
Products
Platform
Developer Tools
Portal

FAQ

Last updated on

  • General questions

    How can I connect to the Dremio Open Catalog?

    Make sure you have configured the Open Catalog on your Dremio instance, refer to Dremio’s official Configure the Open Catalog documentation.

    The catalog endpoint URLs follow the pattern https://catalog.<DREMIO_INSTANCE_ID>.dremio.<STACKIT_REGION>.onstackit.cloud/, where DREMIO_INSTANCE_ID is the UUID of your Dremio instance, and STACKIT_REGION is the region of your STACKIT project (usually eu01). You can also find the catalog URL on the STACKIT Portal, on the “Dremio > URLs” section.

    Different authentication methods are supported, but for Machine-to-Machine applications a Dremio service user is recommended. Refer to Dremio’s official OAuth Client Credentials documentation for more details about generating OAuth client credentials for a new service user.

    You can generate a new access token by submitting an authentication request to the https://<DREMIO_INSTANCE_ID>.dremio.<STACKIT_REGION>.onstackit.cloud/oauth/token endpoint of your Dremio instance. This access token must be provided as a bearer token in your requests to the Catalog.

    I see high memory usage and hash aggregation spilling on my executor, how can I fix it?

    When executors reach their defined memory limit, Dremio spills the data to disk. This usually happens when:

    • Run memory-intensive hash aggregation queries (GROUP BY queries) that process large datasets.
    • Working with memory limited environments.

    You can prevent performance degradation by either optimizing the query leading to high memory consumption, or by changing the engine configuration to give the executors more memory.

    For more details on how to edit engine settings, refer to Dremio’s official Editing the Engine Settings documentation.

    Why is my engine failing to start?

    In July 2026, STACKIT introduced its own flavors for Dremio engines. They are listed in Compute resources for Dremio. Legacy engines or configurations that aren’t referenced in the current offering, are not supported and must be migrated. Old engines won’t be able to get into the Running state and will stay in the Starting state until failure.

    Migrating engines can be achieved by editing them or by recreating them with new settings.

    If you have trouble starting migrated engines, check out other articles in the Dremio FAQ. If you encounter an undocumented issue with your Dremio instance or engine migration, open a support ticket in your STACKIT Portal.

    Why isn't my Dremio engine being used?

    Generally, the life cycle of a SQL query in Dremio is: SQL submitted -> Engine routes -> Queue -> Engine. A newly created engine is neither associated by default to a queue nor to a rule.

    You must own the ADMIN role to be able to make the necessary modifications.

    Log in to your Dremio instance, and head over to the settings section. If your query isn’t starting or arriving at your expected engine, start by looking at the existing routes and verify if the source of your query is being routed to the expected queue.

    Next, check the queue menu and inspect if: _ the engine is associated with any queue at all _ or that the queue where your query being routed to, has your engine associated with it

    More information about Dremio queues and engine routing can be found in the official Dremio documentation: