OpenMetadata
Search…
Connector Configuration
This page provides details on shared configuration settings for connectors.
OpenMetadata connectors require a configuration file with a number of fields to specify settings for the service, data profiler, data filters, sample data, DBT, and security. See below for a simple example of a connector file.
1
{
2
"source": {
3
"type": "redshift",
4
"config": {
5
"host_port": "cluster.name.region.redshift.amazonaws.com:5439",
6
"username": "username",
7
"password": "strong_password",
8
"service_name": "aws_redshift",
9
"data_profiler_enabled": "false",
10
"table_filter_pattern": {
11
"excludes": ["[\\w]*event_vw.*"]
12
},
13
"schema_filter_pattern": {
14
"excludes": ["information_schema.*"]
15
}
16
}
17
},
18
"sink": {
19
"type": "metadata-rest",
20
"config": {}
21
},
22
"metadata_server": {
23
"type": "metadata-server",
24
"config": {
25
"api_endpoint": "http://localhost:8585/api",
26
"auth_provider_type": "no-auth"
27
}
28
}
29
}
Copied!
In the sections below we describe all configuration fields and their settings.
Note: The source.config field in the configuration JSON will include the majority of the settings for your connector. We reference this field through the service connector documentation.

Service Settings

Use service settings to configure your connector to read from the desired service and, optionally, database.

host_port

Use source.config.host_port to send the endpoint for your data service. Use the host:port format illustrated in the example below.
1
"host_port": "cluster.name.region.redshift.amazonaws.com:5439"
Copied!
Please ensure your service is reachable from the host you are using to run metadata ingestion.

username

Edit the value for source.config.username to identify your service user.
1
"username": "username"
Copied!
Note: The user specified should be authorized to read all databases you want to include in the metadata ingestion workflow.

password

Edit the value for source.config.password with the password for your service user.
1
"password": "strong_password"
Copied!

service_name

OpenMetadata uniquely identifies services by their service_name. Edit the value for source.config.service_name with a name that distinguishes this deployment from other services from which you ingest metadata.
1
"service_name": "aws_redshift"
Copied!

database (optional)

If you want to limit metadata ingestion to a single database, include the source.config.database field in your configuration file. If this field is not included, the connector will ingest metadata from all databases the specified user is authorized to read.
To specify a single database to ingest metadata from, provide the name of the database as the value for the source.config.database key as illustrated in the example below.
1
"database": "warehouse"
Copied!
If you want to ingest metadata from two or more databases in a services but not all databases, use the schema_filter_pattern described below to match databases by name using regular expressions or define different workflows using separate config files for each database.

Data Profiler Settings

The data profiler ingests usage information for tables. This enables you to assess frequency of use, reliability, and other details.

data_profiler_enabled

When enabled, the data profiler will run as part of metadata ingestion. Running the data profiler increases the amount of time metadata ingestion requires, but provides the benefits described above.
You may disable the data profiler by including the following field in the source.config object of your configuration file.
1
"data_profiler_enabled": "false"
Copied!
If you want to enable the data profiler, update your configuration file as follows.
1
"data_profiler_enabled": "true"
Copied!
Note: The data profiler is enabled by default if no setting is provided for data_profiler_enabled.

data_profiler_offset (optional)

Use source.config.data_profiler_offset to specify the row offset at which the profiler should begin scanning each table. See below for an example.
1
"data_profiler_offset": "1000"
Copied!
Note: The key source.config.data_profiler_offset value is set to "0" by default.
Note: The source.config.data_profiler_offset field will be removed in a future release of OpenMetadata.

data_profiler_limit (optional)

Use source.config.data_profiler_limit to specify the row limit at which the profiler should conclude scanning each table. You may specify the profiler row limit by including a key-value pair such as the following in the source.config field of your configuration file.
1
"data_profiler_limit": "50000"
Copied!
Note: The value for source.config.data_profiler_limit is set to 50000 by default.
Note: The source.config.data_profiler_limit field will be removed in a future release of OpenMetadata.

Data Filter Settings

include_views (optional)

Use source.config.include_views to control whether or not to include views as part of metadata ingestion and data profiling.
Explicitly include views by adding the following key-value pair in the source.config field of your configuration file.
1
"include_views": "true"
Copied!
Exclude views as follows.
1
"include_views": "false"
Copied!
Note: source.config.include_views is set to true by default.

include_tables (optional)

Use source.config.include_tables to control whether or not to include tables as part of metadata ingestion and data profiling.
Explicitly include tables by adding the following key-value pair in the source.config field of your configuration file.
1
"include_tables": "true"
Copied!
Exclude tables as follows.
1
"include_tables": "false"
Copied!
Note: source.config.include_tables is set to true by default.

table_filter_pattern (optional)

Use source.config.table_filter_pattern to select tables for metadata ingestion by name.
Use source.config.table_filter_pattern.excludes to exclude all tables with names matching one or more of the supplied regular expressions. All other tables will be included. See below for an example. This example is also included in the configuration template provided.
1
"table_filter_pattern": {
2
"excludes": ["information_schema.*", "[\\w]*event_vw.*"]
3
}
Copied!
Use source.config.table_filter_pattern.includes to include all tables with names matching one or more of the supplied regular expressions. All other tables will be excluded. See below for an example.
1
"table_filter_pattern": {
2
"includes": ["corp.*", "dept.*"]
3
}
Copied!
See the documentation for the Python re module for information on how to construct regular expressions.
You may use either excludes or includes but not both in table_filter_pattern.

schema_filter_pattern (optional)

Use source.config.schema_filter_pattern.excludes and source.config.schema_filter_pattern.includes field to select schemas for metadata ingestion by name. The configuration template provides an example.
The syntax and semantics for schema_filter_pattern are the same as for table_filter_pattern. Please see that section for details on use.

Sample Data Settings

generate_sample_data (optional)

Use the source.config.generate_sample_data field to control whether or not to generate sample data to include in table views in the OpenMetadata user interface. See the figure below for an example.
Explicitly include sample data by adding the following key-value pair in the source.config field of your configuration file.
1
"generate_sample_data": "true"
Copied!
If set to true, the connector will collect the first 50 rows of data from each table included in ingestion and catalog that data as sample data to which users can refer in the OpenMetadata user interface.
You can exclude collection of sample data by adding the following key-value pair in the source.config field of your configuration file.
1
"generate_sample_data": "false"
Copied!
Note: generate_sample_data is set to true by default.

DBT Settings

DBT provides transformation logic that creates tables and views from raw data. OpenMetadata includes an integration for DBT that enables you to see the models used to generate a table from that table's details page in the OpenMetadata user interface. See the figure below for an example.
To include DBT models and metadata in your ingestion workflows, specify the location of the DBT manifest and catalog files as fields in your configuration file.

dbt_manifest_file (optional)

Use the field source.config.dbt_manifest_file to specify the location of your DBT manifest file. See below for an example.
1
"dbt_manifest_file": "./dbt/manifest.json"
Copied!

dbt_catalog_file (optional)

Use the field source.config.dbt_catalog_file to specify the location of your DBT catalog file. See below for an example.
1
"dbt_catalog_file": "./dbt/catalog.json"
Copied!
Last modified 22h ago