Prerequisites

Alation Cloud Service Applies to Alation Cloud Service instances of Alation

Customer Managed Applies to customer-managed instances of Alation

Enhanced Connector Enhanced connectors add extended capabilities and require a separate entitlement in addition to your Alation platform license.

Provide Credentials

The Universal DS connector does not require authentication when you load a CSV file directly, because the file is not secured.

The connector requires authentication details when you load a file over SSH: SSH Host, SSH Port, SSH Username, SSH Password, Metadata File Path, and Query Log File Path.

The connector requires authentication details when you load a file from an S3 bucket: Region, Bucket Name, AWS Access Key ID, AWS Secret Access Key, Metadata File Path, and Query Log File Path.

The connector requires authentication details when you load a file from an Azure Storage container: Azure Storage Account Name, Azure Service Principal Auth, Access Key/Shared Access Signature, Azure Storage Container, Azure Storage Endpoint Suffix, Metadata File Path, and Query Log File Path.

Install the Alation Agent

If you are an Alation Cloud Service customer, ensure that Alation Agent is installed and running. The Alation Agent is mandatory for this connector in ACS deployments.

Install the Alation Connector Manager

If you are an on-premise or self-managed customer, ensure that the Alation Connector Manager (ACM) is installed and running. ACM is mandatory for OCF connectors.

Generate the Metadata CSV File

Generate a metadata CSV file that describes your catalogs, schemas, tables, and columns, then make it available to the connector through direct upload in the Alation UI, SSH, AWS S3, or Azure Blob Storage.

Format the Metadata CSV File

The connector reads the metadata CSV as a column-based file. Each row describes one column of one table. The connector maps each field by its position in the row, not by the header name, so the field order must match the default layout below. You can adjust the position of any field on the Metadata Extraction tab if your file uses a different order.

The recommended default layout uses the following header row:

CATALOG,SCHEMA,TABLE_NAME,TABLE_TYPE,TYPE_NAME,DATA_TYPE,COLUMN_NAME,ORDINAL_POSITION,IS_NULLABLE,SCHEMA_REMARKS,TABLE_REMARKS,COLUMN_REMARKS,COLUMN_DEFAULT,PRIMARY_KEY,FOREIGN_KEY

The following table describes each field, its default position, and whether it is mandatory.

Field

Position

Mandatory

Description

CATALOG

1

No

Catalog or database name. When provided, the schema is nested under this catalog.

SCHEMA

2

Yes

Schema name. This column must be present in every row, or the row is skipped.

TABLE_NAME

3

Yes

Table name. This column must be present in every row, or the row is skipped.

TABLE_TYPE

4

No

Table type, for example BASE TABLE or VIEW.

TYPE_NAME

5

No

Source column type name, for example VARCHAR(765) or TIMESTAMP_TZ.

DATA_TYPE

6

No

Normalized column data type, for example TEXT or TIMESTAMP_TZ.

COLUMN_NAME

7

Yes

Column name. This column must be present in every row, or the row is skipped.

ORDINAL_POSITION

8

No

Position of the column within the table, as an integer.

IS_NULLABLE

9

No

Whether the column is nullable. Use YES for nullable columns.

SCHEMA_REMARKS

10

No

Description applied to the schema.

TABLE_REMARKS

11

No

Description applied to the table.

COLUMN_REMARKS

12

No

Description applied to the column.

COLUMN_DEFAULT

13

No

Default value defined for the column.

PRIMARY_KEY

14

No

Whether the column is a primary key. Use YES to mark a primary key.

FOREIGN_KEY

15

No

Referenced column for a foreign key, in schema.table.column or catalog.schema.table.column form.

If a value is not available for an optional field, leave the field empty. For optional numeric fields, enter -1 when the value is not available. If a value contains a comma or double quote, wrap the value in double quotes. Escape a double quote as "".

The table below shows the value each field holds in a sample data row. In the CSV file, these values appear as a single comma-separated line in the same order as the header.

Field

Example Value

CATALOG

FIVETRAN_DATABASE

SCHEMA

SALESFORCE

TABLE_NAME

PARTY_CONSENT

TABLE_TYPE

BASE TABLE

TYPE_NAME

TIMESTAMP_TZ

DATA_TYPE

TIMESTAMP_TZ

COLUMN_NAME

_FIVETRAN_SYNCED

ORDINAL_POSITION

20

IS_NULLABLE

YES

SCHEMA_REMARKS

(empty)

TABLE_REMARKS

(empty)

COLUMN_REMARKS

(empty)

COLUMN_DEFAULT

NULL

PRIMARY_KEY

NO

FOREIGN_KEY

(empty)