Skip to main content

​
Templates and what they expect

  • REST + S3
    • Required: catalog type rest, REST URI, warehouse path (for example, s3://warehouse/wh/)
    • S3 props: io-impl, s3.endpoint, s3.access-key-id, s3.secret-access-key, s3.region, s3.path-style-access
  • JDBC + S3
    • JDBC URI and credentials for the catalog store, S3 warehouse path and S3 props
  • S3 Tables
    • REST endpoint for S3 Tables, S3 region/signing settings, warehouse path, AWS credentials or IAM role
  • Glue + S3
    • Glue region and credentials or role, warehouse path, S3 props
  • Custom
    • Free-form key/value pairs for advanced or mixed setups
  • Load from nimtable config.yaml
    • Reads pre-provisioned catalogs from the mounted config file
  • Load from Spark CLI parameters or spark-defaults.conf
    • Reuses existing Spark Iceberg config to speed onboarding

​
Where definitions live

  • Catalogs added in the UI are stored in Nimtable’s internal database
  • Catalogs in config.yaml are loaded at startup when mounted to /app/config.yaml

​
AWS authentication options

  • Environment variables: AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, AWS_REGION
  • Mounted shared credentials file: ~/.aws/credentials into the container
  • IAM role assumption if supported by your environment (for example, Glue)

​
Tips for RisingWave users

  • Point RisingWave to the Nimtable REST endpoint using catalog.type = 'rest' and catalog.uri
  • Keep storage credentials consistent between Nimtable and RisingWave or enable vended credentials if your REST catalog stack supports them