2022-07-11 16:54:40 +02:00
|
|
|
---
|
|
|
|
title: Run Glue Connector using the CLI
|
|
|
|
slug: /openmetadata/connectors/pipeline/glue/cli
|
|
|
|
---
|
|
|
|
|
|
|
|
<ConnectorIntro connector="Glue" goal="CLI"/>
|
|
|
|
|
|
|
|
<Requirements />
|
|
|
|
|
2022-08-18 11:43:45 +02:00
|
|
|
<PythonMod connector="Glue" module="glue" />
|
|
|
|
|
2022-07-27 20:50:56 +05:30
|
|
|
<MetadataIngestionServiceDev service="pipeline" connector="GluePipeline" goal="CLI"/>
|
2022-07-11 16:54:40 +02:00
|
|
|
|
|
|
|
<h4>Source Configuration - Service Connection</h4>
|
|
|
|
|
|
|
|
- **awsAccessKeyId**: Enter your secure access key ID for your Glue connection. The specified key ID should be
|
|
|
|
authorized to read all databases you want to include in the metadata ingestion workflow.
|
|
|
|
- **awsSecretAccessKey**: Enter the Secret Access Key (the passcode key pair to the key ID from above).
|
|
|
|
- **awsRegion**: Enter the location of the amazon cluster that your data and account are associated with.
|
|
|
|
- **awsSessionToken**: The AWS session token is an optional parameter. If you want, enter the details of your temporary
|
|
|
|
session token.
|
|
|
|
- **endPointURL**: Your Glue connector will automatically determine the AWS Glue endpoint URL based on the region. You
|
|
|
|
may override this behavior by entering a value to the endpoint URL.
|
|
|
|
|
|
|
|
<MetadataIngestionConfig service="pipeline" connector="Glue" goal="CLI" />
|