I have a PostgreSQL database testdb (hosted on 10.55.55.55) as my data source, where I have read-only access. I also have another PostgreSQL database cubejs running on the same server where Cube.js is deployed. Both Cube.js and the DB cubejs are running in Docker.

My .env file contains:

CUBEJS_API_SECRET=secretkey123
CUBEJS_DB_TYPE=postgres
CUBEJS_DB_HOST=10.55.55.55
CUBEJS_DB_PORT=5432
CUBEJS_DB_NAME=testdb
CUBEJS_DB_USER=user_reader
CUBEJS_DB_PASS=password123
CUBEJS_DB_AGGREGATIONS_HOST=postgres
CUBEJS_DB_AGGREGATIONS_PORT=5432
CUBEJS_DB_AGGREGATIONS_NAME=cubejs
CUBEJS_DB_AGGREGATIONS_USER=cubejs_user
CUBEJS_DB_AGGREGATIONS_PASS=cubejs_password123
CUBEJS_PRE_AGGREGATIONS_SCHEMA=cube_preaggs
CUBEJS_DEV_MODE=true
CUBEJS_EXTERNAL_DEFAULT=true

I'm making the following POST request:

curl -X POST http://localhost:4000/cubejs-api/v1/load   \
  -H "Content-Type: application/json"   \
  -H "Authorization: secretkey123"   \
  --data '{
    "query": {
      "measures": [],
      "dimensions": [
        "customers.name",
        "customers.email"
      ],
      "timeDimensions": [
        {
          "dimension": "customers.created_at",
          "granularity": "day"
        }
      ],
      "limit": 10
    }
  }'

The request returns the expected results, but the cube_preaggs schema is not created in the cubejs database (I've created the cube_preaggs schema beforehand, but the table for storing the sql-query still doesn’t appear).

How can I ensure that pre-aggregations are stored in the cubejs database? (In the future, we plan to use ClickHouse for pre-aggregations.) I see in the logs that the SQL query for pre-aggregation is executed in cube_preaggs without errors. Did I specify the database correctly in the .env file? What should I check?

I’d appreciate any insights, thanks in advance!