> ## Documentation Index
> Fetch the complete documentation index at: https://docs.open-metadata.org/llms.txt
> Use this file to discover all available pages before exploring further.

# Run the Databricks Connector Externally

> Use YAML to ingest Databricks metadata including notebook lineage, schema, and profiling.

![Databricks](https://docs.open-metadata.org/public/images/connectors/databrick.webp)

# Databricks

PROD

> [!TIP]
>
> As per the [documentation](https://docs.databricks.com/en/data-governance/unity-catalog/tags.html#manage-tags-with-sql-commands) here, note that we only support metadata `tag` extraction for Databricks version 13.3 version and higher.

In this section, we provide guides and references to use the Databricks connector.

> [!NOTE]
>
> **Supported Authentication Types:**
>
> * **Personal Access Token** — Token-based workspace authentication generated from User Settings in Databricks
> * **Databricks OAuth** — OAuth2 Machine-to-Machine authentication using Service Principal credentials
> * **Azure AD Setup** — Azure Active Directory authentication using Azure Service Principal (for Azure Databricks workspaces)

Configure and schedule Databricks metadata and profiler workflows from the OpenMetadata UI:

* [Requirements](#requirements)
* [Metadata Ingestion](#metadata-ingestion)
* [Query Usage](#query-usage)
* [Lineage](#lineage)
* [Data Profiler](#data-profiler)
* [Data Quality](#data-quality)
* [dbt Integration](#dbt-integration)

## How to Run the Connector Externally

To run the Ingestion via the UI you'll need to use the OpenMetadata Ingestion Container, which comes shipped with
custom Airflow plugins to handle the workflow deployment.

If, instead, you want to manage your workflows externally on your preferred orchestrator, you can check
the following docs to run the Ingestion Framework **anywhere**.

> [card] [External Schedulers](https://docs.open-metadata.org/v1.13.x/deployment/ingestion.md)
>
> Get more information about running the Ingestion Framework Externally

## Requirements

### Python Requirements

Use a Python version supported by the `openmetadata-ingestion` package that matches your OpenMetadata server. To find the supported range for your release, check the `Requires-Python` metadata of that release’s ingestion package.

To run the Databricks ingestion, you will need to install:

```bash theme={null}
pip3 install "openmetadata-ingestion[databricks]"
```

### Permission Requirement

To enable full functionality of metadata extraction, profiling, usage, and lineage features in OpenMetadata, the following permissions must be granted to the relevant users in your Databricks environment.

### Metadata and Profiling Permissions

These permissions are required on the catalogs, schemas, and tables from which metadata and profiling information will be ingested.

```sql theme={null}
GRANT USE CATALOG ON CATALOG <catalog_name> TO `<user>`;
GRANT USE SCHEMA ON SCHEMA <schema_name> TO `<user>`;
GRANT SELECT ON TABLE <table_name> TO `<user>`;
```

Ensure these grants are applied to all relevant tables for metadata ingestion and profiling operations.

### Usage and Lineage

These permissions enable OpenMetadata to extract query history and construct lineage information.

```sql theme={null}
-- Query history for usage analytics and SQL-based lineage
GRANT SELECT ON SYSTEM.QUERY.HISTORY TO `<user>`;
GRANT USE SCHEMA ON SCHEMA system.query TO `<user>`;

-- System lineage tables for table and column-level lineage
GRANT SELECT ON system.access.table_lineage TO `<user>`;
GRANT SELECT ON system.access.column_lineage TO `<user>`;
GRANT USE SCHEMA ON SCHEMA system.access TO `<user>`;
```

These permissions allow access to Databricks system tables that track query activity and lineage relationships, enabling lineage and usage statistics generation.

### View Definitions

To extract view definitions, the user needs access to the information schema:

```sql theme={null}
GRANT SELECT ON INFORMATION_SCHEMA.VIEWS TO `<user>`;
```

### Tags (Databricks 13.3+)

To extract Databricks tags on catalogs, schemas, tables, and columns, the following permissions are required:

```sql theme={null}
GRANT SELECT ON `<catalog_name>`.information_schema.catalog_tags TO `<user>`;
GRANT SELECT ON `<catalog_name>`.information_schema.schema_tags TO `<user>`;
GRANT SELECT ON `<catalog_name>`.information_schema.table_tags TO `<user>`;
GRANT SELECT ON `<catalog_name>`.information_schema.column_tags TO `<user>`;
```

> [!WARNING]
>
> Tag extraction requires Databricks Runtime 13.3 or higher. If your cluster is running an older version, tags will not be extracted.

> [!TIP]
>
> Adjust \<user>, \<catalog\_name>, \<schema\_name>, and \<table\_name> according to your specific deployment and security requirements.

## Metadata Ingestion

All connectors are defined as JSON Schemas.
[Here](https://github.com/open-metadata/OpenMetadata/blob/main/openmetadata-spec/src/main/resources/json/schema/entity/services/connections/database/databricksConnection.json)
you can find the structure to create a connection to Databricks.
In order to create and run a Metadata Ingestion workflow, we will follow
the steps to create a YAML configuration able to connect to the source,
process the Entities if needed, and reach the OpenMetadata server.
The workflow is modeled around the following
[JSON Schema](https://github.com/open-metadata/OpenMetadata/blob/main/openmetadata-spec/src/main/resources/json/schema/metadataIngestion/workflow.json)

### 1. Define the YAML Config

#### Source Configuration

Configure the source type and service name for your Databricks connector.

#### Catalog

**catalog**: Catalog of the data source (Example: hive\_metastore). This is optional parameter, if you would like to restrict the metadata reading to a single catalog. When left blank, OpenMetadata Ingestion attempts to scan all the catalog.

#### Database Schema

**databaseSchema**: DatabaseSchema of the data source. This is optional parameter, if you would like to restrict the metadata reading to a single databaseSchema. When left blank, OpenMetadata Ingestion attempts to scan all the databaseSchema.

#### Authentication Type

**authType**: Choose one of the following authentication methods:

* **Personal Access Token**: Provide `token` — generated from User Settings → Developer → Access Tokens in your Databricks workspace.
* **Databricks OAuth**: Provide `clientId` and `clientSecret` for a Service Principal created in your Databricks Account Console.
* **Azure AD Setup**: Provide `azureClientId`, `azureClientSecret`, and `azureTenantId` for an Azure Service Principal registered in Azure Active Directory.

#### Host Port

**hostPort**: Enter the fully qualified hostname and port number for your Databricks deployment in the Host and Port field.

#### HTTP Path

**httpPath**: Databricks compute resources URL.

#### Connection Timeout

**connectionTimeout**: The maximum amount of time (in seconds) to wait for a successful connection to the data source. If the connection attempt takes longer than this timeout period, an error will be returned.

#### Query History Table

**queryHistoryTable**: Table name to fetch the query history from. Defaults to `system.query.history`.

#### Connection Options

**Connection Options (Optional)**: Enter the details for any additional connection options that can be sent to database during the connection. These details must be added as Key-Value pairs.

#### Connection Arguments

**Connection Arguments (Optional)**: Enter the details for any additional connection arguments such as security or protocol configs that can be sent to database during the connection. These details must be added as Key-Value pairs.

* In case you are using Single-Sign-On (SSO) for authentication, add the `authenticator` details in the Connection Arguments as a Key-Value pair as follows: `"authenticator" : "sso_login_url"`

#### Source Config

The `sourceConfig` is defined [here](https://github.com/open-metadata/OpenMetadata/blob/main/openmetadata-spec/src/main/resources/json/schema/metadataIngestion/databaseServiceMetadataPipeline.json):

* **markDeletedTables**: To flag tables as soft-deleted if they are not present anymore in the source system.
* **markDeletedStoredProcedures**: Optional configuration to soft delete stored procedures in OpenMetadata if the source stored procedures are deleted. Also, if the stored procedure is deleted, all the associated entities like lineage, etc., with that stored procedure will be deleted.
* **markDeletedSchemas**: Optional configuration to soft delete schemas stored in OpenMetadata if the source schema is deleted. Setting this flag to true will only keep filtered schemas and delete any other schemas that do not match schemaFilterPattern or do not exist at source.
* **markDeletedDatabases**: Additional optional configuration for soft deletion, providing a granular option to select which particular entities should be deleted.
* **includeTables**: Set to `true` or `false` to ingest table data. Default is `true`.
* **includeViews**: Set to `true` or `false` to ingest view definitions.
* **includeTags**: Optional configuration to toggle the tags ingestion.
* **includeOwners**: Set the **Include Owners** toggle to control whether to include owners to the ingested entity if the owner email matches with a user stored in the OM server as part of metadata ingestion. If the ingested entity already exists and has an owner, the owner will not be overwritten.
* **ownerConfig**: Advanced, hierarchical configuration for assigning owners to databases, schemas, and tables during ingestion, with per-entity mappings and inheritance. For more information, see [Hierarchical Owner Configuration](https://docs.open-metadata.org/connectors/ingestion/workflows/metadata/hierarchical-owner-configuration.md).
* **includeStoredProcedures**: Optional configuration to toggle the Stored Procedures ingestion.
* **includeDDL**: Optional configuration to toggle the DDL Statements ingestion.
* **overrideMetadata** *(boolean)*: Set the **Override Metadata** toggle to control whether to override the existing metadata in the OpenMetadata server with the metadata fetched from the source. If the toggle is set to true, the metadata fetched from the source will override the existing metadata in the OpenMetadata server. If the toggle is set to false, the metadata fetched from the source will not override the existing metadata in the OpenMetadata server. This is applicable for fields like description, tags, owner, and displayName.
* **queryLogDuration**: Configuration to tune how far we want to look back in query logs to process Stored Procedures results.
* **queryParsingTimeoutLimit**: Configuration to set the timeout for parsing the query in seconds.
* **useFqnForFiltering**: Regex will be applied on the fully qualified name (for example, `service_name.db_name.schema_name.table_name`) instead of the raw name (for example, `table_name`).
* **databaseFilterPattern**, **schemaFilterPattern**: Note that the filter supports regex as include or exclude. You can find examples [here](https://docs.open-metadata.org/connectors/ingestion/workflows/metadata/filter-patterns/database.md).
* **tableFilterPattern**: Note that the filter supports regex as include or exclude. You can find examples [here](https://docs.open-metadata.org/connectors/ingestion/workflows/metadata/filter-patterns/table.md).
* **threads (beta)**: The number of threads to use when extracting the metadata using multithreading.
* **databaseMetadataConfigType** *(string)*: Database Source Config Metadata Pipeline type.
* **incremental (beta)**: Incremental Extraction configuration. Currently implemented for [BigQuery](https://docs.open-metadata.org/connectors/ingestion/workflows/metadata/incremental-extraction/bigquery.md), [Redshift](https://docs.open-metadata.org/connectors/ingestion/workflows/metadata/incremental-extraction/redshift.md), and [Snowflake](https://docs.open-metadata.org/connectors/ingestion/workflows/metadata/incremental-extraction/snowflake.md).

#### Sink Configuration

To send the metadata to OpenMetadata, it needs to be specified as `type: metadata-rest`.

#### Workflow Configuration

The main property here is `openMetadataServerConfig`, where you can define the host and security provider of your OpenMetadata installation.

* **loggerLevel**: Specify the logger level depending on your needs. If you are troubleshooting an ingestion, use `DEBUG` for more detailed traces.
* **JWT token**: JWT tokens allow clients to authenticate against the OpenMetadata server. See [Enable JWT Tokens](https://docs.open-metadata.org/deployment/security/enable-jwt-tokens.md) and [JWT Troubleshooting](https://docs.open-metadata.org/deployment/security/jwt-troubleshooting.md) for more information.
* **storeServiceConnection**: If set to `true` (default), sensitive information is stored encrypted with the Fernet Key or externally if you have configured a [Secrets Manager](https://docs.open-metadata.org/deployment/secrets-manager.md). If set to `false`, the service is created, but the service connection information is only used by the Ingestion Framework at runtime and is not sent to the OpenMetadata server.
* **SSL configuration**: If you have added SSL to the [OpenMetadata server](https://docs.open-metadata.org/deployment/security/enable-ssl.md), configure the certificates for ingestion. Set `verifySSL` to `ignore`, or set it to `validate` and provide `sslConfig.caCertificate` with a local path to the server certificate. See [SSL Troubleshooting](https://docs.open-metadata.org/deployment/security/enable-ssl/ssl-troubleshooting.md) for more information.
* **ingestionPipelineFQN**: Fully qualified name of the ingestion pipeline, used to identify the current ingestion pipeline.

databricks\_config.yaml
Copy full code

```yaml theme={null}
source:
  type: databricks
  serviceName: local_databricks
  serviceConnection:
    config:
      type: Databricks
      catalog: hive_metastore
      databaseSchema: default
      authType:
        token: <databricks token>  # Personal Access Token
      # -- OR for Databricks OAuth --
      # authType:
      #   clientId: <service principal client id>
      #   clientSecret: <service principal client secret>
      # -- OR for Azure AD --
      # authType:
      #   azureClientId: <azure client id>
      #   azureClientSecret: <azure client secret>
      #   azureTenantId: <azure tenant id>
      hostPort: <databricks connection host & port>  # REQUIRED
      httpPath: <http path of databricks cluster>  # REQUIRED
      connectionTimeout: 120
      queryHistoryTable: system.query.history
      # connectionOptions:
      #   key: value
      # connectionArguments:
      #   key: value
```

```yaml theme={null}
  sourceConfig:
    config:
      type: DatabaseMetadata
      markDeletedTables: true
      markDeletedStoredProcedures: true
      markDeletedSchemas: true
      markDeletedDatabases: true
      includeTables: true
      includeViews: true
      # includeTags: true
      # includeOwners: false
      # ownerConfig:
      #   default: data-platform-team
      #   database:
      #     database1: owner1
      #   databaseSchema:
      #     database1.schema1: owner1
      #   table:
      #     database1.schema1.table1: owner1
      #   enableInheritance: true
      # includeStoredProcedures: true
      # includeDDL: true
      # overrideMetadata: false
      # queryLogDuration: 1
      # queryParsingTimeoutLimit: 300
      # useFqnForFiltering: false
      # threads: 1
      # databaseMetadataConfigType: ()
      # incremental:
      #   enabled: true
      #   lookbackDays: 7
      #   safetyMarginDays: 1
      # databaseFilterPattern:
      #   includes:
      #     - database1
      #     - database2
      #   excludes:
      #     - database3
      #     - database4
      # schemaFilterPattern:
      #   includes:
      #     - schema1
      #     - schema2
      #   excludes:
      #     - schema3
      #     - schema4
      # tableFilterPattern:
      #   includes:
      #     - users
      #     - type_test
      #   excludes:
      #     - table3
      #     - table4
```

```yaml theme={null}
sink:
  type: metadata-rest
  config: {}
```

```yaml theme={null}
workflowConfig:
  loggerLevel: INFO  # DEBUG, INFO, WARNING or ERROR
  openMetadataServerConfig:
    hostPort: "http://localhost:8585/api"
    authProvider: openmetadata
    securityConfig:
      jwtToken: "{bot_jwt_token}"
    ## Store the service Connection information
    storeServiceConnection: true  # false
    ## Secrets Manager Configuration
    # secretsManagerProvider: aws, azure or noop
    # secretsManagerLoader: airflow or env
    ## If SSL, fill the following
    # verifySSL: validate  # or ignore
    # sslConfig:
    #   caCertificate: /local/path/to/certificate
# ingestionPipelineFQN: <service name>.<ingestion name> ## e.g., "my_redshift.metadata"
```

### 2. Run with the CLI

First, we will need to save the YAML file. Afterward, and with all requirements installed, we can run:

```bash theme={null}
metadata ingest -c <path-to-yaml>
```

Note that from connector to connector, this recipe will always be the same. By updating the YAML configuration,
you will be able to extract metadata from different sources.

## Query Usage

The Query Usage workflow will be using the `query-parser` processor.

After running a Metadata Ingestion workflow, we can run Query Usage workflow.
While the `serviceName` will be the same to that was used in Metadata Ingestion, so the ingestion bot can get the `serviceConnection` details from the server.

### 1. Define the YAML Config

This is a sample config for databricks Usage:

#### Source Configuration

Configure the source type and service name for your usage workflow.

#### Usage Config Type

**type**: Set to `DatabaseUsage` for database usage ingestion.

#### Query Log Duration

**queryLogDuration**: Configuration to tune how far we want to look back in query logs to process usage data (in days).

#### Stage File Location

**stageFileLocation**: Temporary file name to store the query logs before processing. Absolute file path required.

Note that the location is a directory that will be cleaned at the end of the ingestion.

#### Result Limit

**resultLimit**: Configuration to set the limit for query logs.

#### Query Log File Path

**queryLogFilePath**: Configuration to set the file path for query logs. If instead of getting the query logs from the database we want to pass a file with the queries.

#### Processor Configuration

Choose the `query-parser` processor to parse and process the query logs.

#### Stage Configuration

Configure the staging location for table usage data before it's sent to OpenMetadata.

#### Bulk Sink Configuration

Configure the bulk sink for metadata usage ingestion.

\{connector}\_usage.yaml
Copy full code

```yaml theme={null}
source:
  type: databricks-usage
  serviceName: databricks
  sourceConfig:
    config:
      type: DatabaseUsage
      # Number of days to look back
      queryLogDuration: 7
      # This is a directory that will be DELETED after the usage runs
      stageFileLocation: <path to store the stage file>
      # resultLimit: 1000
      # If instead of getting the query logs from the database we want to pass a file with the queries
      # queryLogFilePath: path-to-file
processor:
  type: query-parser
  config: {}
stage:
  type: table-usage
  config:
    filename: /tmp/databricks_usage
bulkSink:
  type: metadata-usage
  config:
    filename: /tmp/databricks_usage
```

### 2. Run with the CLI

After saving the YAML config, we will run the command the same way we did for the metadata ingestion:

```bash theme={null}
metadata usage -c <path-to-yaml>
```

## Lineage

After running a Metadata Ingestion workflow, we can run Lineage workflow.
While the `serviceName` will be the same to that was used in Metadata Ingestion, so the ingestion bot can get the `serviceConnection` details from the server.

### 1. Define the YAML Config

This is a sample config for databricks Lineage:

#### Source Configuration

Configure the source type and service name for your lineage workflow.

You can find all the definitions and types for the `sourceConfig` [here](https://github.com/open-metadata/OpenMetadata/blob/main/openmetadata-spec/src/main/resources/json/schema/metadataIngestion/databaseServiceQueryLineagePipeline.json).

#### Lineage Config Type

**type**: Set to `DatabaseLineage` for database lineage ingestion.

#### Query Log Duration

**queryLogDuration**: Configuration to tune how far we want to look back in query logs to process lineage data in days.

#### Parsing Timeout Limit

**parsingTimeoutLimit**: Configuration to set the timeout for parsing the query in seconds.

#### Filter Condition

**filterCondition**: Condition to filter the query history.

#### Result Limit

**resultLimit**: Configuration to set the limit for query logs.

#### Query Log File Path

**queryLogFilePath**: Configuration to set the file path for query logs. If instead of getting the query logs from the database we want to pass a file with the queries.

#### Database Filter Pattern

**databaseFilterPattern**: Regex to only fetch databases that matches the pattern.

#### Schema Filter Pattern

**schemaFilterPattern**: Regex to only fetch tables or databases that matches the pattern.

#### Table Filter Pattern

**tableFilterPattern**: Regex to only fetch tables or databases that matches the pattern.

#### Override View Lineage

**overrideViewLineage**: Set the 'Override View Lineage' toggle to control whether to override the existing view lineage.

#### Process View Lineage

**processViewLineage**: Set the 'Process View Lineage' toggle to control whether to process view lineage.

#### Process Query Lineage

**processQueryLineage**: Set the 'Process Query Lineage' toggle to control whether to process query lineage.

#### Process Stored Procedure Lineage

**processStoredProcedureLineage**: Set the 'Process Stored ProcedureLog Lineage' toggle to control whether to process stored procedure lineage.

#### Threads

**threads**: Number of Threads to use in order to parallelize lineage ingestion.

#### Sink Configuration

To send the metadata to OpenMetadata, it needs to be specified as `type: metadata-rest`.

\{connector}\_lineage.yaml
Copy full code

```yaml theme={null}
source:
  type: databricks-lineage
  serviceName: databricks
  sourceConfig:
    config:
      type: DatabaseLineage
      # Number of days to look back
      queryLogDuration: 1
      parsingTimeoutLimit: 300
      # filterCondition: query_text not ilike '--- metabase query %'
      resultLimit: 1000
      # If instead of getting the query logs from the database we want to pass a file with the queries
      # queryLogFilePath: /tmp/query_log/file_path
      # databaseFilterPattern:
      #   includes:
      #     - database1
      #     - database2
      #   excludes:
      #     - database3
      # schemaFilterPattern:
      #   includes:
      #     - schema1
      #     - schema2
      #   excludes:
      #     - schema3
      # tableFilterPattern:
      #   includes:
      #     - table1
      #     - table2
      #   excludes:
      #     - table3
      #     - table4
      overrideViewLineage: false
      processViewLineage: true
      processQueryLineage: true
      processStoredProcedureLineage: true
      threads: 1
sink:
  type: metadata-rest
  config: {}
```

* You can learn more about how to configure and run the Lineage Workflow to extract Lineage data from [here](https://docs.open-metadata.org/connectors/ingestion/workflows/lineage.md)

### 2. Run with the CLI

After saving the YAML config, we will run the command the same way we did for the metadata ingestion:

```bash theme={null}
metadata ingest -c <path-to-yaml>
```

## Data Profiler

The Data Profiler workflow will be using the `orm-profiler` processor.

After running a Metadata Ingestion workflow, we can run the Data Profiler workflow.
While the `serviceName` will be the same to that was used in Metadata Ingestion, so the ingestion bot can get the `serviceConnection` details from the server.

### 1. Define the YAML Config

This is a sample config for the profiler:

#### Source Configuration

Configure the source type and service name for your profiler workflow.

#### Profiler Config Type

**type**: Set to `Profiler` for data profiling ingestion.

#### Profile Sample Config

**profileSampleConfig**: How much data the profiler and tests run on.

* **sampleConfigType**: `DYNAMIC` (default) resolves the sample size at runtime from the table's row count. `STATIC` uses a fixed size you set yourself.
* **config**: the settings for the chosen type. With `DYNAMIC`, `smartSampling: true` applies the built-in tiers. Set it to `false` and provide `thresholds` to define your own. With `STATIC`, set `profileSample`, and optionally `profileSampleType` and `samplingMethodType`.

#### Thread Count

**threadCount**: Number of threads to use during metric computations.

#### Timeout Seconds

**timeoutSeconds**: Profiler Timeout in Seconds.

#### Database Filter Pattern

**databaseFilterPattern**: Regex to only fetch databases that matches the pattern.

#### Schema Filter Pattern

**schemaFilterPattern**: Regex to only fetch tables or databases that matches the pattern.

#### Table Filter Pattern

**tableFilterPattern**: Regex to only fetch tables or databases that matches the pattern.

#### Processor Configuration

Choose the `orm-profiler`. Its config can also be updated to define tests from the YAML itself instead of the UI.

**tableConfig**: `tableConfig` allows you to set up some configuration at the table level including:

* Profile sample settings per table
* Custom profile queries
* Column-level configuration (include/exclude columns, specific metrics)
* Partition configuration for large tables

#### Sink Configuration

To send the metadata to OpenMetadata, it needs to be specified as `type: metadata-rest`.

\{connector}\_profiler.yaml
Copy full code

```yaml theme={null}
source:
  type: databricks
  serviceName: databricks
  sourceConfig:
    config:
      type: Profiler
      # profileSampleConfig:
      #   sampleConfigType: DYNAMIC  # DYNAMIC (default) or STATIC
      #   config:
      #     smartSampling: true
      # threadCount: 5
      # timeoutSeconds: 43200
      # databaseFilterPattern:
      #   includes:
      #     - database1
      #     - database2
      #   excludes:
      #     - database3
      # schemaFilterPattern:
      #   includes:
      #     - schema1
      #     - schema2
      #   excludes:
      #     - schema3
      # tableFilterPattern:
      #   includes:
      #     - table1
      #     - table2
      #   excludes:
      #     - table3
processor:
  type: orm-profiler
  config: {}  # Remove braces if adding properties
    # tableConfig:
    #   - fullyQualifiedName: <table fqn>
    #     profileSampleConfig:
    #       sampleConfigType: STATIC  # omit entirely to inherit dynamic sampling
    #       config:
    #         profileSample: <number between 0 and 99>
    #     profileQuery: <query to use for sampling data for the profiler>
    #     columnConfig:
    #       excludeColumns:
    #         - <column name>
    #       includeColumns:
    #         - columnName: <column name>
    #         - metrics:
    #           - MEAN
    #           - MEDIAN
    #           - ...
    #     partitionConfig:
    #       enablePartitioning: <set to true to use partitioning>
    #       partitionColumnName: <partition column name>
    #       partitionIntervalType: <TIME-UNIT, INTEGER-RANGE, INGESTION-TIME, COLUMN-VALUE>
    #       Pick one of the variation shown below
    #       ----'TIME-UNIT' or 'INGESTION-TIME'-------
    #       partitionInterval: <partition interval>
    #       partitionIntervalUnit: <YEAR, MONTH, DAY, HOUR>
    #       ------------'INTEGER-RANGE'---------------
    #       partitionIntegerRangeStart: <integer>
    #       partitionIntegerRangeEnd: <integer>
    #       -----------'COLUMN-VALUE'----------------
    #       partitionValues:
    #         - <value>
    #         - <value>
sink:
  type: metadata-rest
  config: {}
```

* You can learn more about how to configure and run the Profiler Workflow to extract Profiler data and execute the Data Quality from [here](https://docs.open-metadata.org/how-to-guides/data-quality-observability/profiler/profiler-workflow.md)

### 2. Run with the CLI

After saving the YAML config, we will run the command the same way we did for the metadata ingestion:

```bash theme={null}
metadata profile -c <path-to-yaml>
```

Note now instead of running `ingest`, we are using the `profile` command to select the Profiler workflow.

> [card] [Data Profiler](https://docs.open-metadata.org/v1.13.x/how-to-guides/data-quality-observability/profiler/profiler-workflow.md)
>
> Find more information about the Data Profiler here

## Auto Classification

The Auto Classification workflow will be using the `orm-profiler` processor.

After running a Metadata Ingestion workflow, we can run the Auto Classification workflow.
While the `serviceName` will be the same to that was used in Metadata Ingestion, so the ingestion bot can get the `serviceConnection` details from the server.

### 1. Define the YAML Config

This is a sample config for the Auto Classification Workflow:

#### Source Configuration

Configure the source type and service name for your auto classification workflow.

#### Auto Classification Config Type

**type**: Set to `AutoClassification` for automatic PII tagging.

#### Store Sample Data

**storeSampleData**: Option to turn on/off storing sample data. If enabled, we will ingest sample data for each table.

#### Enable Auto Classification

**enableAutoClassification**: Optional configuration to automatically tag columns that might contain sensitive information.

#### Confidence

**confidence**: Set the Confidence value for which you want the column to be tagged as PII. Confidence value ranges from 0 to 100. A higher number will yield less false positives but more false negatives. A lower number will yield more false positives but less false negatives.

#### Database Filter Pattern

**databaseFilterPattern**: Regex to only fetch databases that matches the pattern.

#### Schema Filter Pattern

**schemaFilterPattern**: Regex to only fetch tables or databases that matches the pattern.

#### Table Filter Pattern

**tableFilterPattern**: Regex to only fetch tables or databases that matches the pattern.

#### Processor Configuration

Choose the `orm-profiler`. Its config can also be updated to define tests from the YAML itself instead of the UI.

**tableConfig**: `tableConfig` allows you to set up some configuration at the table level.

#### Sink Configuration

To send the metadata to OpenMetadata, it needs to be specified as `type: metadata-rest`.

\{connector}\_auto\_classification.yaml
Copy full code

```yaml theme={null}
source:
  type: databricks
  serviceName: databricks
  sourceConfig:
    config:
      type: AutoClassification
      # storeSampleData: true
      # enableAutoClassification: true
      # confidence: 80
      # databaseFilterPattern:
      #   includes:
      #     - database1
      #     - database2
      #   excludes:
      #     - database3
      # schemaFilterPattern:
      #   includes:
      #     - schema1
      #     - schema2
      #   excludes:
      #     - schema3
      # tableFilterPattern:
      #   includes:
      #     - table1
      #     - table2
      #   excludes:
      #     - table3
processor:
  type: orm-profiler
  config: {}
sink:
  type: metadata-rest
  config: {}
```

### 2. Run with the CLI

After saving the YAML config, we will run the command the same way we did for the metadata ingestion:

```bash theme={null}
metadata classify -c <path-to-yaml>
```

> [!TIP]
>
> Now instead of running `ingest`, we are using the `classify` command to select the Auto Classification workflow.

## Data Quality

### Adding Data Quality Test Cases from yaml config

When creating a JSON config for a test workflow the source configuration is very simple.

```yaml theme={null}
source:
  type: TestSuite
  serviceName: <your_service_name>
  sourceConfig:
    config:
      type: TestSuite
      entityFullyQualifiedName: <entityFqn>
```

The only sections you need to modify here are the `serviceName` (this name needs to be unique) and `entityFullyQualifiedName` (the entity for which we'll be executing tests against) keys.

Once you have defined your source configuration you'll need to define te processor configuration.

```yaml theme={null}
processor:
  type: "orm-test-runner"
  config:
    forceUpdate: <false|true>
    testCases:
      - name: <testCaseName>
        testDefinitionName: columnValueLengthsToBeBetween
        columnName: <columnName>
        parameterValues:
          - name: minLength
            value: 10
          - name: maxLength
            value: 25
      - name: <testCaseName>
        testDefinitionName: tableRowCountToEqual
        parameterValues:
          - name: value
            value: 10
```

The processor type should be set to ` "orm-test-runner"`. For accepted test definition names and parameter value names refer to the [tests page](https://docs.open-metadata.org/how-to-guides/data-quality-observability/quality/tests-yaml.md).

> [!TIP]
>
> Note that while you can define tests directly in this YAML configuration, running the
> workflow will execute ALL THE TESTS present in the table, regardless of what you are defining in the YAML.
>
> This makes it easy for any user to contribute tests via the UI, while maintaining the test execution external.

You can keep your YAML config as simple as follows if the table already has tests.

```yaml theme={null}
processor:
  type: "orm-test-runner"
  config: {}
```

### Key reference:

* `forceUpdate`: if the test case exists (base on the test case name) for the entity, implements the strategy to follow when running the test (i.e. whether or not to update parameters)
* `testCases`: list of test cases to add to the entity referenced. Note that we will execute all the tests present in the Table.
* `name`: test case name
* `testDefinitionName`: test definition
* `columnName`: only applies to column test. The name of the column to run the test against
* `parameterValues`: parameter values of the test

The `sink` and `workflowConfig` will have the same settings as the ingestion and profiler workflow.

### Full  `yaml` config example

```yaml theme={null}
source:
  type: TestSuite
  serviceName: MyAwesomeTestSuite
  sourceConfig:
    config:
      type: TestSuite
      entityFullyQualifiedName: MySQL.default.openmetadata_db.tag_usage
#     testCases: ["run_only_this_test_case"] # Optional, if not provided all tests will be executed

processor:
  type: "orm-test-runner"
  config:
    forceUpdate: false
    testCases:
      - name: column_value_length_tagFQN
        testDefinitionName: columnValueLengthsToBeBetween
        columnName: tagFQN
        parameterValues:
          - name: minLength
            value: 10
          - name: maxLength
            value: 25
      - name: table_row_count_test
        testDefinitionName: tableRowCountToEqual
        parameterValues:
          - name: value
            value: 10

sink:
  type: metadata-rest
  config: {}
workflowConfig:
  openMetadataServerConfig:
    hostPort: <OpenMetadata host and port>
    authProvider: <OpenMetadata auth provider>
```

### How to Run Tests

To run the tests from the CLI execute the following command

```
metadata test -c /path/to/my/config.yaml
```

## dbt Integration

You can learn more about how to ingest dbt models' definitions and their lineage [here](https://docs.open-metadata.org/v1.13.x/connectors/database/dbt.md).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.