Working with Data Governance Tools

You can configure Data Collector to integrate with data governance tools, giving you visibility into data movement - where the data came from, where it’s going to, and who is interacting with it.

Data governance tool integration is available with a paid subscription. For details, contact IBM.

Data Collector can work with the following data governance tools:

Apache Atlas: If you have Apache Atlas installed, you can configure Data Collector to publish metadata about running pipelines to Apache Atlas. When publishing metadata, Data Collector uses Apache Kafka to send the metadata to Apache Atlas. The data is sent in JSON format.
Cloudera Navigator: If you use Cloudera Manager and have Cloudera Navigator installed, you can configure Data Collector to publish metadata about running pipelines to Cloudera Navigator. When publishing metadata, Data Collector makes an HTTPS request to Cloudera Navigator using basic authentication. The data is sent in JSON format.
Collibra: If you use Collibra Platform, you can configure Data Collector to publish metadata about running pipelines to Collibra. When publishing metadata, Data Collector passes data to Collibra using the Collibra import API. The data is sent in JSON format.

You then use one of the data governance tools to explore the pipeline metadata, including viewing lineage diagrams of the metadata.

Data Collector uses separate threads to publish metadata to data governance tools - so enabling metadata publishing has minimal effect on the running pipeline threads. Every pipeline publishes metadata when the pipeline starts and when it stops. Some origins and destinations publish metadata only once, typically when they initialize. However, some stages publish metadata each time they create a new object - for example, when the Hadoop FS or Local FS destination creates a new output file.

Supported Stages

At this time, only some pipeline stages support publishing metadata to data governance tools.

Data Collector publishes metadata for all supported stages in all running pipelines. If you have multiple Data Collectors that run pipelines, configure each Data Collector to publish metadata to the same governance tool instance.

Data Collector can publish metadata for the following stages:

Amazon S3 origin
Dev Data Generator origin
Directory origin
JDBC Query Consumer origin
Kafka Multitopic Consumer origin
Kinesis Consumer origin
SFTP/FTP/FTPS Client origin
Hadoop FS destination
Local FS destination
HBase destination
Kafka Producer destination
Kudu destination

Note: Remember that you don't enable metadata publishing at the stage level or even at the pipeline level. You simply configure Data Collector to publish metadata for all running pipelines.

When a pipeline includes an unsupported stage, the data governance tool does not display that stage as an input or output in the lineage diagram. For example, if a running pipeline includes an unsupported origin and the Hadoop FS destination, then the data governance tool displays the pipeline as having 0 inputs and multiple outputs - one output for each generated output file. If a running pipeline includes no supported origins or destinations, the data governance tool displays the pipeline as having 0 inputs and 0 outputs.

Apache Atlas

You can configure Data Collector to publish metadata about running pipelines to Apache Atlas. You then use Apache Atlas to explore the pipeline metadata, including viewing lineage diagrams of the metadata.

Data governance tool integration is available with a paid subscription. For details, contact IBM.

To integrate Data Collector with Apache Atlas, download and install one of the following Apache Atlas stage libraries as a custom stage library:


Stage Library	Description
streamsets-datacollector-apache-atlas_0_8_1-lib	Use when working with HDP version 2.6.x and Apache Atlas 0.8.1.
streamsets-datacollector-apache-atlas_1_1_0-lib	Use when working with HDP version 3.1.x and Apache Atlas 1.1.0.

You then configure Data Collector to connect to Apache Atlas. You also must deploy the IBM StreamSets data model to the machine where the Apache Atlas server is installed.

Prerequisites

Before you enable Data Collector to publish pipeline metadata to Apache Atlas, you must complete the following prerequisites:

Verify that Data Collector has a required Hortonworks Data Platform (HDP) stage library installed

To connect to and publish metadata to Apache Atlas, Data Collector requires HDP versions 2.6.x or 3.1.x.

To verify that Data Collector has a valid HDP stage library installed, click the Package Manager icon (

) to display the list of installed stage libraries. If a required library version is not installed, install the library before configuring Data Collector to publish pipeline metadata to Apache Atlas.

Configure the required settings in Apache Atlas

To connect to and publish metadata to Apache Atlas, you must configure the following required settings in Apache Atlas:

Configure Apache Atlas to use HBase as a persistence engine.
Configure Apache Atlas to use Apache Kafka for notifications.
Ensure that the graphing engine is enabled for Apache Atlas.

For more information on configuring Apache Atlas, see the Apache Atlas documentation.

Step 1. Download and Install the Stage Library

Download and install the appropriate Apache Atlas stage library as a custom stage library.

Using the link provided by email, download the tarball containing the Apache Atlas stage library.
Extract the tarball.
The tarball is extracted into the following directory:
```
<stage library name>/lib/
```
For example, the Apache Atlas version 0.8.1 stage library is extracted to:
```
streamsets-datacollector-apache-atlas_0_8_1-lib/lib/
```
The Apache Atlas version 1.1.0 stage library is extracted to:
```
streamsets-datacollector-apache-atlas_1_1_0-lib/lib/
```
Include the stage library as a custom stage library in an external resource archive for the deployment.

Step 2. Configure the Connection to Apache Atlas

Configuring Data Collector to connect to Apache Atlas involves working with the following files:

Data Collector configuration properties: Add lineage publisher properties to the Data Collector configuration properties.
Apache Atlas application properties file: If Data Collector is installed on a separate machine from the Apache Atlas server, copy the atlas-application.properties file to the Data Collector machine. Modify the Apache Atlas application properties file as needed, and then specify the location of the file in the Java configuration properties of the deployment.

Warning: If metadata publishing is enabled and Apache Atlas cannot access the atlas-application.properties file at Data Collector start up, then Data Collector fails to start.

To enable metadata publishing to Apache Atlas, in Control Hub, edit the deployment. In the Configure Engine section, click Advanced Configuration. Then, click Data Collector Configuration.

Add lineage publisher properties to the configuration properties.

For Apache Atlas 0.8.1, add the following lines to the file:

# Lineage Publisher
#
lineage.publishers=atlas
lineage.publisher.atlas.def=streamsets-datacollector-apache-atlas_0_8_1-lib::com_streamsets_pipeline_stage_plugin_atlas_AtlasLineagePublisher

For Apache Atlas 1.1.0, add the following lines to the file:

# Lineage Publisher
#
lineage.publishers=atlas
lineage.publisher.atlas.def=streamsets-datacollector-apache-atlas_1_1_0-lib::com_streamsets_pipeline_stage_plugin_atlas_AtlasLineagePublisher1_1_0

If Data Collector is installed on a separate machine from the Apache Atlas server, copy the atlas-application.properties file to the Data Collector machine.

Store the atlas-application.properties file in the Data Collector configuration directory.
Verify that the atlas-application.properties file is correctly configured.
For example, if you copied the atlas-application.properties file to the Data Collector machine, you might need to modify the following properties which specify the Kafka installation on the Apache Atlas server:
```
atlas.kafka.zookeeper.connect=localhost:9026
atlas.kafka.bootstrap.servers=localhost:9027
```
For more information on the properties in this file, see the Apache Atlas documentation.
Define the -Datlas.conf system property in the Java configuration properties of the deployment. In Control Hub, edit the deployment. In the Configure Engine section, click Advanced Configuration. Then, click Java Configuration.
The -Datlas.conf system property defines the location of the local atlas-application.properties file. Define the Atlas system property in the Java Options property as follows:
```
-Datlas.conf=<configuration directory>
```
The defined configuration directory must end in a forward slash (/).
Save the changes to the deployment and restart all engine instances.

Step 3. Deploy the IBM StreamSets Data Model to Apache Atlas

Download the IBM StreamSets data model and deploy the data model to the Apache Atlas server.

Using the link provided by email, download the data model for Apache Atlas:
```
0090-streamsets_model.json
```
Deploy the data model to the machine where the Apache Atlas server is installed.
On the Apache Atlas server, copy the data model to the same directory that contains the Sqoop data model.
For example, if using Hortonworks version 2.6.1 for testing, copy the data model to the following directory:
```
/usr/hdp/2.6.1.0-129/atlas/models/0090-streamsets_model.json
```
Restart the Apache Atlas server so that the server can recognize the newly deployed data model.

Viewing Published Metadata

Data Collector publishes pipeline metadata to Apache Atlas when a pipeline stops. Apache Atlas lists each pipeline by pipeline title and displays supported origins as inputs and supported destinations as outputs. Apache Atlas also includes other metadata, such as the pipeline description, version, labels, and the user who started the pipeline.

For example, let's assume that you run a pipeline that includes a JDBC Query Consumer origin and a Local FS destination. Apache Atlas displays a single input representing the JDBC Query Consumer origin. As the pipeline runs, the Local FS destination creates multiple output files. Apache Atlas displays multiple outputs, each output representing one of the generated output files.

Apache Atlas displays the details of the published pipeline metadata as follows. Note how our example pipeline is listed with 1 input and 3 outputs - one output for each generated output file:

Apache Atlas displays the pipeline lineage diagram as follows:

Cloudera Navigator

You can configure Data Collector to publish metadata about running pipelines to Cloudera Navigator. You then use Cloudera Navigator to explore the pipeline metadata, including viewing lineage diagrams of the metadata.

Data governance tool integration is available with a paid subscription. For details, contact IBM.

To integrate Data Collector with Cloudera Navigator, download and install one of the following Cloudera Navigator stage libraries as a custom stage library:


Stage Library	Description
streamsets-datacollector-cm_5_10-lib	Use when working with Cloudera Manager version 5.10.
streamsets-datacollector-cm_5_11-lib	Use when working with Cloudera Manager version 5.11.
streamsets-datacollector-cm_5_12-lib	Use when working with Cloudera Manager version 5.12.
streamsets-datacollector-cm_5_13-lib	Use when working with Cloudera Manager version 5.13.
streamsets-datacollector-cm_5_14-lib	Use when working with Cloudera Manager version 5.14.
streamsets-datacollector-cm_5_15-lib	Use when working with Cloudera Manager version 5.15.
streamsets-datacollector-cm_6_1-lib	Use when working with Cloudera Manager version 6.1.

You then configure Data Collector to connect to Cloudera Navigator.

Prerequisites

Before you enable Data Collector to publish pipeline metadata to Cloudera Navigator, complete the following prerequisites:

Verify that Data Collector has a required Cloudera CDH stage library version installed

To connect to and publish metadata to Cloudera Navigator, Data Collector requires Cloudera Manager versions 5.10 through 5.15 or version 6.1.

To verify that Data Collector has a valid Cloudera CDH stage library installed, click the Package Manager icon (

) to display the list of installed stage libraries. If a required library version is not installed, install the library before configuring Data Collector to publish pipeline metadata to Cloudera Navigator.

Add the Cloudera Navigator data management roles to your Cloudera Manager cluster

Add the following Navigator roles to the Cloudera Management Service:

Cloudera Navigator Audit Server role
Cloudera Navigator Metadata Server role

For instructions, see the Cloudera Manager documentation.

Step 1. Download and Install the Stage Library

Download and install the Cloudera Navigator stage library as a custom stage library.

Using the link provided by email, download the tarball containing the Cloudera Navigator stage library for your version of Cloudera Manager.
For example, if Data Collector has the Cloudera CDH stage library version 5.14 installed, download the streamsets-datacollector-cm_5_14-lib-1.6.0.tgz file.
Extract the tarball.
The tarball is extracted into the following directory:
```
<stage library name>/lib/
```
For example:
```
streamsets-datacollector-cm_5_14-lib/lib/
```
Include the stage library as a custom stage library in an external resource archive for the deployment.

Step 2. Configure the Connection to Cloudera Navigator

Add lineage publisher properties to the Data Collector configuration properties, and then configure the properties as needed.

Important: Instead of entering sensitive data such as passwords in clear text in the configuration properties, you can protect the sensitive data by storing the data in an external location and then using functions to retrieve the data.

Configuring the properties includes enabling metadata publishing and defining the connection to Cloudera Navigator.

If Cloudera Navigator is configured for TLS/SSL, configuring the properties also requires specifying the local truststore file that Data Collector uses to verify the identity of the Cloudera Navigator Metadata Server.

Warning: If metadata publishing is enabled and Data Collector cannot connect to Cloudera Navigator at start up, then Data Collector fails to start.

To enable metadata publishing, in Control Hub, edit the deployment. In the Configure Engine section, click Advanced Configuration. Then, click Data Collector Configuration.

Add the following lineage publisher properties to the configuration properties:

# Lineage Publisher
#
lineage.publishers=navigator
lineage.publisher.navigator.def=streamsets-datacollector-cm_5_14-lib::com_streamsets_pipeline_stage_plugin_navigator_NavigatorLineagePublisher
lineage.publisher.navigator.config.application_url=http://localhost:18630
lineage.publisher.navigator.config.navigator_url=http://navigatorhost:7187
lineage.publisher.navigator.config.namespace=sdc
lineage.publisher.navigator.config.username=user
lineage.publisher.navigator.config.password=passwd
lineage.publisher.navigator.config.autocommit=false

To define the connection to Cloudera Navigator, modify the default values of the following properties:


Metadata Publisher Property	Description
lineage.publishers	Application to publish metadata to. Use the default value, `navigator`.
lineage.publisher.navigator.def	Stage library name used to connect to Cloudera Navigator, entered in the following format: `<stage library name>::com_streamsets_pipeline_ stage_plugin_navigator_NavigatorLineagePublisher` Define the appropriate Cloudera CDH stage library name to enable metadata publishing. For example, if you downloaded version 6.1 of the stage library, define the stage library name as follows: `streamsets-datacollector-cm_6_1-lib` Do not modify the second half of the property value.
lineage.publisher.navigator. config.application_url	URL to Data Collector. For example: `http://<hostname>:18630/`
lineage.publisher.navigator. config.navigator_url	URL to the Cloudera Navigator UI. For example: `http://<Navigator Metadata Server host>:<port number>/` `<Navigator Metadata Server host>` is the name of the host on which you are running the Navigator Metadata Server role. `<port number>` is the port configured for the role.
lineage.publisher.navigator. config.namespace	Namespace that Cloudera Navigator uses to uniquely identify Data Collector properties.
lineage.publisher.navigator. config.username	User name to connect to Cloudera Navigator.
lineage.publisher.navigator. config.password	Password for the Cloudera Navigator account. To protect the password, store the password in an external location and then use a function to retrieve the password.
lineage.publisher.navigator. config.autocommit	Enables Cloudera Navigator to immediately process the published metadata. In most cases, use the default value of `false`. Setting to true can use a large number of resources on the Cloudera Navigator machine.

If Cloudera Navigator is configured for TLS/SSL, add the following lines to the Lineage Publisher section of the Data Collector configuration properties and then modify the default values of the properties:

lineage.publisher.navigator.config.ssl_truststore_location=truststore.jks
lineage.publisher.navigator.config.ssl_truststore_password=changeit


Metadata Publisher Property for TLS/SSL	Description
lineage.publisher.navigator. config.ssl_truststore_location	Path to the truststore file containing the Cloudera Navigator Metadata Server certificate. Enter an absolute path to the file or a path relative to the Data Collector resources directory: `$SDC_RESOURCES`.
lineage.publisher.navigator. config.ssl_truststore_password	Password to the truststore file. To protect the password, store the password in an external location and then use a function to retrieve the password.

Restart Data Collector for the changed properties to take effect.

Viewing Published Metadata

You can view published pipeline metadata in near real time in Cloudera Navigator. Cloudera Navigator lists each pipeline by pipeline title and displays supported origins as inputs and supported destinations as outputs. Cloudera Navigator also includes a URL to the pipeline within Data Collector along with other metadata, such as the pipeline ID, version, labels, and the user who started the pipeline.

For example, let's assume that you run a pipeline that includes a JDBC Query Consumer origin and a Local FS destination. Cloudera Navigator displays a single input representing the JDBC Query Consumer origin. As the pipeline runs, the Local FS destination creates multiple output files. Cloudera Navigator displays multiple outputs, each output representing one of the generated output files.

Cloudera Navigator displays the details of the published pipeline metadata as follows. Note how our example Jdbc2LocalFS pipeline is listed with 1 input and 42 outputs - one output for each generated output file:

Cloudera Navigator displays the pipeline lineage diagram as follows:

Collibra

You can configure Data Collector to publish metadata about running pipelines to Collibra. You then use Collibra to explore the pipeline metadata, including viewing lineage diagrams of the metadata. You can configure Data Collector to publish metadata to the latest version of Collibra.

Data governance tool integration is available with a paid subscription. For details, contact IBM.

To integrate Data Collector with Collibra, download and install the following Collibra stage library as a custom stage library:


Stage Library	Description
streamsets-datacollector-collibra-lib	Use to work with Collibra Platform.

You then configure Data Collector to connect to Collibra. You also must deploy the IBM StreamSets data model to the machine where the Collibra server is installed.

Prerequisites

Before you enable Data Collector to publish pipeline metadata to Collibra, complete the following prerequisites.

This document includes steps for performing tasks in Collibra for your convenience. For the latest steps, see the Collibra documentation.

You must be a Collibra system administrator to complete these tasks:

Import IBM StreamSets Collibra resources

To connect to Collibra from Data Collector, you must download and import the Collibra resource file.

If necessary, place a request with customer support for access to the Collibra resource file, streamsets-collibra.cma.
In the Collibra user interface, from your Collibra instance, click Settings > Migration > Import, then upload the streamsets-collibra.cma file.

Create a community in the Collibra instance

To connect to Collibra from Data Collector, you must create a community in Collibra for Data Collector, and assign the Community Manager role to a user account. You use the Community Manager user account to configure the connection to Collibra.

In the Collibra user interface, from your Collibra instance, click Create in the top menu bar.
In the Create Community dialog box, specify a name to uniquely identify the community, such as Data Collector, then click Create.
in the Add Responsibility dialog box, specify a user for the Community Manager role, and click Add.

Step 1. Download and Install the Stage Library

Download and install the Collibra stage library as a custom stage library.

Using the link provided by email, download the tarball containing the Collibra stage library.
Extract the tarball.
The tarball is extracted into the following directory:
```
<stage library name>/lib/
```
For example, the Collibra version 1.0.0 stage library is extracted to:
```
streamsets-datacollector-collibra-lib/lib/
```
Include the stage library as a custom stage library in an external resource archive for the deployment.

Step 2. Configure the Connection to Collibra

Configuring Data Collector to connect to Collibra involves adding lineage publisher properties to the Data Collector configuration properties.

To enable metadata publishing to Collibra, in Control Hub, edit the deployment. In the Configure Engine section, click Advanced Configuration. Then, click Data Collector Configuration.

Add the following lineage publisher properties to the configuration properties:

# Lineage Publisher
lineage.publishers=collibra
lineage.publisher.collibra.def=streamsets-datacollector-collibra-lib::com_streamsets_pipeline_stage_plugin_collibra_CollibraLineagePublisher
lineage.publisher.collibra.config.collibra_url=<collibra url>
lineage.publisher.collibra.config.community=<community name>
lineage.publisher.collibra.config.username=<user name>
lineage.publisher.collibra.config.password=<password>

To define the connection to Collibra, configure the properties as follows:


Metadata Publisher Property	Description
lineage.publishers	Application to publish metadata to. Use the default value: `collibra`.
lineage.publisher.collibra.def	Stage library name used to connect to Collibra. Use the default value: `streamsets-datacollector-collibra-lib::com_streamsets_pipeline_stage_plugin_collibra_CollibraLineagePublisher`
lineage.publisher.collibra.config.collibra_url	URL for your Collibra instance. Use the following format: `https://<collibra host>:<collibra port>`
lineage.publisher.collibra.config.community	Community that Collibra uses to uniquely identify Data Collector properties. Specify the community name that you created in the prerequisite tasks.
lineage.publisher.collibra.config.username	User name to connect to Collibra. Specify the user account that you assigned to the Community Manager role in the prerequisite tasks.
lineage.publisher.collibra.config.password	Password for the Collibra account.

Restart Data Collector.

Viewing Published Metadata

Data Collector publishes pipeline metadata to Collibra when a pipeline stops. Collibra lists each pipeline by pipeline title and displays supported origins as inputs and supported destinations as outputs. Collibra also includes other metadata, such as the pipeline ID, version, and the user who started the pipeline.

For example, say you run a pipeline that includes an Oracle Bulkloader origin and a Local FS destination. Collibra displays a single input representing the origin. As the pipeline runs, the Local FS destination creates multiple output files. Collibra displays multiple outputs, each output representing one of the generated output files.

Collibra displays the details of the published pipeline metadata as follows. Note how the pipeline is listed with 1 input and 4 outputs - one output for each generated output file:

Collibra displays the pipeline lineage diagram as follows: