Gary K
03/09/2026, 5:49 AMError executing SQL Query: ERROR: relation "dataplane" does not exist ?kapa.ai
03/09/2026, 5:50 AMdataplane table (and likely related tables like dataplane_group) does not exist in your Airbyte database, which means the required schema migrations have not been applied correctly.
Based on the knowledge sources, here are the most likely causes and solutions:
Likely Causes
1. Incomplete or failed database migration during an upgrade: When upgrading Airbyte (especially to versions like 1.6.0, 1.8.0, or 2.0.0), database migrations that create or alter the dataplane and dataplane_group tables may fail partway through. For example, a known bug in v1.6.0 involved a migration (V1_1_1_016__AddDataplaneGroupIdToConnection) that could leave dataplane_group_id as NULL for some connections, causing downstream errors. [migration bug]
2. Stale or mismatched docker-compose.yml / Helm values: If your configuration files (e.g., docker-compose.yml or Helm values.yaml) are out of date relative to the Airbyte version you're running, migrations may not execute properly. One user fixed a similar startup failure by ensuring their docker-compose.yml and environment variables (e.g., RUN_DATABASE_MIGRATION_ON_STARTUP) were up to date. [startup failure fix]
3. Insufficient database user permissions: If the database user Airbyte connects with lacks the necessary privileges, it may be unable to create or modify tables during migration. [permissions issue]
Suggested Steps
1. Verify your config files are up to date â ensure your docker-compose.yml or Helm chart values match the version of Airbyte you are deploying.
2. Check database user permissions â ensure the Airbyte DB user has full privileges on the Airbyte database:
GRANT ALL PRIVILEGES ON DATABASE airbyte TO <airbyte_user>;
3. Check if migrations ran â inspect the airbyte_configs_migrations table to see which migrations were applied and whether the ones creating the `dataplane`/`dataplane_group` tables are present.
4. If upgrading to v2.0, be aware of a known issue where two dataplane_group records can exist after the upgrade, causing jobs to be routed to the wrong dataplane. A workaround is to update the dataplane_group_id in the workspace table, or to add the following to your Helm config before upgrading: [upgrade issue]
global:
dataplaneGroups:
defaultDataplaneGroupName: "AUTO"
If none of these apply, restoring the database to its pre-upgrade state and retrying the upgrade has been reported as a last resort. [restore workaround] đ 0 button đ 0 button âšī¸ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.