Apache Superset How to add multiple database connections to superset

Bug description

Hi , I am automating the dashboard import process by keeping all the files in Repo and importing them to Superset. I have a requirement to add multiple databases. My current folder structure is below for Reporting database. If I need to add a new database eg, Superset_metadb.yaml how the structure will be for handling multiple databses /tmp/repo/databases/databases.zip databases/ databases/metadata.yaml databases/datasets/ databases/datasets/Reporting/ databases/datasets/Reporting/Inventory_Cyclecount_by_Task_Priority_and_Zone.yaml databases/datasets/Reporting/Inventory_Task_Ended.yaml databases/Reporting.yaml

Screenshots/recordings

No response

Superset version

5.0.0

Python version

3.11

Node version

18 or greater

Browser

Chrome

Additional context

No response

Checklist

• [x] I have searched Superset docs and Slack and didn't find a solution to my problem.
• [x] I have searched the GitHub issue tracker and didn't find a similar bug report.
• [x] I have checked Superset's logs for errors and if I found a relevant Python stacktrace, I included it here as text in the "additional context" section.

Comment From: dosubot[bot]

To import multiple database connections into Superset (v5.0.0) using your automated process, structure your ZIP file like this:

databases/
  metadata.yaml
  Reporting.yaml
  Superset_metadb.yaml
  ...
datasets/
  Reporting/
    Inventory_Cyclecount_by_Task_Priority_and_Zone.yaml
    Inventory_Task_Ended.yaml
  Superset_metadb/
    ... (datasets for this DB)

Each database connection gets its own YAML file under databases/ (e.g., Reporting.yaml, Superset_metadb.yaml). Datasets for each database go in their own subfolder under datasets/ (e.g., datasets/Reporting/, datasets/Superset_metadb/). This structure is explicitly supported by the import logic in Superset 5.x and is the standard approach for bulk importing databases and datasets via the UI or API [source] [source].

When importing, you can provide passwords for each database as a JSON map, keyed by the path to the YAML file (e.g., {"databases/Reporting.yaml": "password1", "databases/Superset_metadb.yaml": "password2"}).

No special naming convention is required beyond placing each database YAML in the databases/ directory and organizing datasets by database name under datasets/. This approach works for automated imports and is robust for handling multiple databases in Superset.

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}    

Comment From: Devika7733

@dosu., Then how do we pass password for multiple databases in API call if I am keeping multiple databses in a zip file like: databases/ metadata.yaml Reporting.yaml Superset_metadb.yaml . I am currently using the below endpoints for uploading Assets: local response=$(curl -k -s -w "%{http_code}" -b cookies.txt -o "${asset_type}_response_body.txt" \ --url "https://superset.test.com/api/v1/database/import/" \ --header 'accept: application/json' \ --header Authorization": Bearer $ACCESS_TOKEN" \ --header 'content-type: multipart/form-data' \ --header "referer: https://superset.test.com" \ --header "x-csrftoken: $CSRF_TOKEN" \ --form "formData=@$zip_file" \ --form overwrite=true \ --form "passwords={\"databases/SuperSet.Metadata.yaml\": \"$SUPERSET_DB_PASS\"}")

Comment From: dosubot[bot]

To pass passwords for multiple databases in your API call, use the passwords form field as a JSON string mapping each database YAML path (as it appears in your zip, e.g., databases/Reporting.yaml, databases/Superset_metadb.yaml) to its password. The keys must match the file paths exactly, including case and directory structure. For example:

--form 'passwords={"databases/Reporting.yaml":"'"$REPORTING_PASS"'", "databases/Superset_metadb.yaml":"'"$SUPERSET_DB_PASS"'"}'

Your curl command should look like:

curl -k -s -w "%{http_code}" -b cookies.txt -o "${asset_type}_response_body.txt" \
  --url "https://superset.test.com/api/v1/database/import/" \
  --header 'accept: application/json' \
  --header "Authorization: Bearer $ACCESS_TOKEN" \
  --header 'content-type: multipart/form-data' \
  --header "referer: https://superset.test.com" \
  --header "x-csrftoken: $CSRF_TOKEN" \
  --form "formData=@$zip_file" \
  --form overwrite=true \
  --form 'passwords={"databases/Reporting.yaml":"'"$REPORTING_PASS"'", "databases/Superset_metadb.yaml":"'"$SUPERSET_DB_PASS"'"}'

If you need to provide SSH tunnel credentials, use the same pattern for the relevant fields (ssh_tunnel_passwords, ssh_tunnel_private_keys, ssh_tunnel_private_key_passwords) as JSON objects keyed by the database YAML path reference.

Common pitfalls: make sure your JSON is properly escaped, and the paths match exactly what’s in your zip file. If a password is missing or a path is incorrect, the import will fail reference.

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}