This is a guide to installing the WhereScape Enablement Pack for Databricks for WhereScape RED.



Prerequisites For PostgreSQL Metadata

Before you begin, ensure to meet the following prerequisites:

  • Create Database and ODBC DSN :
           1. Supported* version of PostgreSQL (PostgreSQL 12 or higher)
                 i.   A database to house the RED Metadata Repository.
                 ii.  A database for the Range Table DB (Optional)
                 iii. A database to house scheduler (Optional)

RED supports the following versions for the metadata repository: PostgreSQL 12 or higher

Software Installations 

  • WhereScape RED10 with a valid license key entered and EULA accepted.
  • WhereScape Enablement Pack for target database version RED10.

Windows Powershell (64 bit) version 4 or higher

To check the Windows Powershell Version:

  • Run below command in Windows Powershell: 
    Get-Host|Select-Object Version
  • Run below command in Command Prompt:
    powershell $psversiontable


Run the following command using PowerShell:


  • The security protocol TLS 1.0 and 1.1 used by PowerShell to communicate with PowerShell gallery has deprecated and TLS 1.2 has been made mandatory:
    1 [Net.ServicePointManager]::SecurityProtocol = [Net.ServicePointManager]::SecurityProtocol -bor [Net.SecurityProtocolType]::Tls12
2 Register-PSRepository -Default -Verbose
3 Set-PSRepository -Name PSGallery -InstallationPolicy Trusted
  • Progress bar placeholder info line:
    1 Install-Module -Name PoshProgressBar -SkipPublisherCheck -Force

Prerequisites For Target Database Databricks

  • Before you begin the following prerequisites must be met:
  • Create Database and ODBC DSN:
  • Software Installations
    • Databricks CLI - Refer to Setup Guide Databricks CLI Setup 
  • Python 3.8 or higher
    • Select Add Python 3.8 to PATH from the installation Window
    • Pip Manager Install with the command: python -m pip install --upgrade pip


Enablement Pack Setup Scripts

The Enablement Pack Install process is entirely driven by scripts. The following table outlines these scripts, their purpose and if Run as Administrator is required. 


ScriptDescriptionRun As AdministratorUse
1Setup_Enablement_Pack.ps1

Setup and configure a RED Metadata Repository for target database If RED repository exists then updates the repository with:

  1. Templates
  2. Scripts
  3. Extended Properties
  4. Datatype Mappings
  5. UI Configurations
YesNew and Existing installations
2install_WslPython_Modules.batInstalls or updates WslPython Modules and required Python libraries on this machine.YesNew and Existing installations
3import_python_templates.ps1Imports or updates the Python Templates to a RED Metadata Repository. Also includes any Script ImportsNo*Existing installations
4set_default_templates.ps1Applies the RED Connection defaults in a RED Metadata Repository for Python or Powershell templates.No*Existing installations


Note

Note that on some systems the execution Windows Powershell scripts is disabled by default, see the troubleshooting section for workarounds.

Each Powershell script in the table above provides some help at the command line, to view this information pass the -help parameter to the script. For example, > .\Setup_Enablement_Pack.ps1 -help  

Step-By-Step Guide

Setup and configure RED Metadata Repository

Run Powershell as Administrator:         

  • Script 1 > Powershell -ExecutionPolicy Bypass -File .\Setup_Enablement_Pack.ps1 

Important Upgrade Notes

If RED repository exists, it will prompt to upgrade the repository.

This enablement pack will overwrite any existing Source Enablement Pack UI Configs:

Connection UI ConfigLoad UI Config
Amazon S3Load From Amazon S3
Azure Data Lake Storage Gen2Load From Azure Data Lake Storage Gen2
Google CloudLoad From Google Cloud

To ensure existing Source Enablement Pack connections and associated Load Tables continue to browse and load, go into UI Configuration Maintenance in RED before installing this Enablement Pack and rename the affected UI Configurations.

While the updated Load Template will work with previous Source Enablement Pack's we recommend moving these previous versions of Load Tables to newly created Parser-based connections following this install. The earlier versions of the Source Enablement Pack will be deprecated following this release.  

Install or Update WhereScape Python Modules      

Run the Script as Administrator         

  • Script 2 > install_WslPython_Modules.bat    

This script performs two tasks:                 

  1. Installs the WhereScape WslPython modules to C:\Program Data\WhereScape\Modules\      
  2. Uses PIP to download or update required Python libraries - for offline install please see the required library list for Python in the Troubleshooting section.

Install or Update WhereScape Python Templates (For Existing Installations)

Run these scripts as Administrator          

  • Script 2 > install_WslPython_Modules.bat          
  • Script 3 > . .\import_python_templates.ps1          
  • Script 4 > . .\set_default_templates.ps1

Note

Skip this step for new installations.

Set Connection defaults for a Template Set (For Existing Installations)

  • Script 4 > . .\set_default_templates.ps1
  • Choose Python when prompted.  

Post Install Steps - Optional

If you used the script Setup_Enablement_Pack.ps1 then the following optional post-install steps are available

Configure Connections

There were Three connections added that will optionally require your attention:

  1. Connection: 'Database Source System' - this connection was set up as an example source connection,
    • open it's properties and set it up for a source DB in your environment
    • or you can remove it if not required.
  2. Connection: 'Databricks' - this connection was set as per parameters provided in script 1
    1. open properties and check if the Database ID is set correctly
    2. open its properties and check the extended properties tab, set it up for HTTP_PATH, SERVER_HOSTNAME, DB_ACCESS_TOKEN, and DBFS_TMP

Enable Script Launcher Toolbar

Several stand-alone scripts provide some features such as "Ranged Loading", these scripts have been added to the Script Launcher menu but you will need to enable the menu toolbar item to see them.
To enable the Script Launcher menu in RED: Select menu item View>Toolbars>Script Launcher

Source Enablement Pack Support

Source Pack Name

Supported by Databricks

Supported Features

Prerequisites

Amazon S3

Yes

Bulk load to Databricks

Include the Access Key and Secret Key in the Amazon S3 Cloud Parser Connection for S3. For guidance on obtaining these credentials, please refer to the relevant documentation: {+}https://docs.aws.amazon.com/IAM/latest/UserGuide/security-creds.html+

Azure Data Lake Storage Gen2

Yes

Bulk load to Databricks

Add the SAS Token to the ADLG2 Cloud Parser Connection. Refer to {+}https://learn.microsoft.com/en-us/azure/storage/common/storage-sas-overview+ for information on SAS Tokens.

Google Cloud Storage

Yes

Bulk load to Databricks

Step 1: Service Account Setup

  1. Create a service account in Google Cloud Console.
  2. Navigate to IAM and Admin > Service Accounts.
  3. Click + CREATE SERVICE ACCOUNT, enter details, and create the account.

    Step 2: Generate Access Key for GCS Bucket
  4. In the service accounts list, click the created account.
  5. In the Keys section, click ADD KEY > Create new key.
  6. Choose JSON key type and click CREATE to download the key file.

    Step 3: Bucket Configuration
  7. Configure bucket details in Google Cloud Console.
  8. Navigate to the Permissions tab and click ADD next to Permissions.
  9. Grant Storage Admin permission to the service account on the bucket.
  10. Click SAVE.

    Step 4: Databricks Cluster Configuration
  11. In the Spark Config tab, set the keys using the following snippet:

    spark.hadoop.google.cloud.auth.service.account.enable true
    spark.hadoop.fs.gs.auth.service.account.email <client-email>
    spark.hadoop.fs.gs.project.id <project-id>
    spark.hadoop.fs.gs.auth.service.account.private.key secrets/scope/gsa_private_key
    spark.hadoop.fs.gs.auth.service.account.private.key.id secrets/scope/gsa_private_key_id
    Replace `<client-email>` and `<project-id>` with values from the downloaded JSON key.
    For detailed documentation, refer to:
    {+}https://learn.microsoft.com/en-us/azure/databricks/storage/gcs+

Windows Parser

1.  CSV
2.  Excel
3.  JSON
4.  XML
5.  AVRO
6.  ORC
7.  PARQUET

Load Template, Source Properties will have the option to select parser type to load the files.

Refer to Windows Parser Guide

Troubleshooting and Tips

Run As Administrator

Press the Windows Key on your keyboard and start typing cmd.exe, when the cmd.exe icon shows up in the search list right click it to bring up the context menu, select Run As Administrator
Now you have an admin prompt navigate to to the folder where you have unpacked your WhereScape Red Enablement Pack using the cd command:
C:\Windows\system32> cd <full path to the unpacked folder> 
Run batch (.bat) scripts from the administrator prompt by simply typing the name at the prompt and clicking enter, for example:
C:\temp\EnablementPack>install_WslPython_Modules.bat
Run Powershell (.ps1) scripts from the administrator prompt by typing the Powershell run script command, for example:
C:\temp\EnablementPack>Powershell -ExecutionPolicy Bypass -File .\Setup_Enablement_Pack.ps1

In the event you can not bypass the Powershell execution policy due to group policies you can instead try "-ExecutionPolicy RemoteSigned" which should allow unsigned local scripts.

Setting Up Databricks Configuration

  1. Add a system variable DATABRICKS_CONFIG_FILE to point to a location that permits you to configure the databricks-cli.

  2. Open the command prompt and configure databricks-cli using databricks configure --aad-token.
  3. On running this command, config file should be created in the location specified in the config file system variable

Windows Powershell Script Execution

On some systems, Windows Powershell script execution is disabled by default. There are several workarounds for this which can be found by searching the term "Powershell Execution Policy".
Here is the most common workaround that WhereScape suggests, which does not permanently change the execution rights:
Start a Windows CMD prompt as Administrator, change the directory to your script directory, and run the WhereScape Powershell scripts with this command:

  • cmd:>Powershell -ExecutionPolicy Bypass -File .\<script_file_name.ps1>

Restarting failed scripts

Some setup scripts will track each step and output the step number when there is a failure. To restart from the failed step (or to skip the step) provide the parameter "-startAtStep <step number>" to the script.
Example: 
Powershell -ExecutionPolicy Bypass -File .\<script_file_name.ps1> -startAtStep 123

To avoid having to provide all the parameters again you can copy the full command line with parameters from the first "INFO" message from the beginning of the console output.

Python requirements for offline install

Additionally to the base Python installation being required, the WhereScape Python Template set also requires certain additional Python libraries. The install scripts use the PIP (package manager) to download these libraries, however, for offline installs, you will need to install the required libraries yourself.
Required Python libraries/add-ons:

  • pywin32-ctypes
  • python-tds
  • pywin32
  • glob2
  • gzip-reader
  • regex
  • pyodbc
  • databricks
  • databricks-sql-connector

If a valid RED installation can not be found

If you have Red 8.5.1.x or higher installed but the script (Setup_Enablement_Pack.ps1) fails to find it on your system then you are most likely running the PowerShell (x86) version which does not show the installed 64-bit apps by default. Please open a 64-bit version of Powershell instead and re-run the script.