Upload a file without creating a table

Use a bucket when the original bytes matter or when the content has no table shape: images, PDFs, archives, models, and folders are examples. A bucket belongs to a catalog and domain.

Before you begin

You need permission to write in the destination domain. Check the domain in Data → Catalog first. If the bucket does not exist, create it from the Catalog Create menu where your role permits it.

Upload and verify

1. Open Data → Ingestion, choose Upload, then choose Bucket as the type.
2. Choose the catalog, domain, and bucket.
3. Select files or a folder. Keep relative paths if the folder structure matters.
4. Upload and inspect the per-file result. A partial result requires you to retry only the failed files.
5. Open the bucket in the catalog. Select one file to see its detail page, path, content type, size, and classification when available.

The file is a file. Converting it to a table is a separate action with its own validation and lineage.

If the upload is refused

Read the exact error. A missing domain, denied permission, duplicate path, or unsupported size needs a different fix. The UI must not turn a refusal into a successful empty bucket.