Cloudera Data Science Workbench Administration Guide

This topic describes how to configure and manage a Cloudera Data Science Workbench deployment as a site administrator. By default, the first user account that signs up for the Cloudera Data Science Workbench becomes a site administrator. Site administrators can manage other users, monitor resources, secure access to the deployment, and upload license keys for the product.
To access the site administrator dashboard:
  1. Go to the Cloudera Data Science Workbench web application (http://cdsw.company.com) and log in as a site administrator.
  2. On the left sidebar, click Admin. You will see an array of tabs for all the tasks you can perform as a site administrator.

The rest of this topic describes some common tasks for a Cloudera Data Science Workbench site administrator.

Managing Users

As a site administrator you can add new users, assign or modify privileges for existing users, and monitor user activity on the Cloudera Data Science Workbench deployment.

Adding New Users

To invite new users, navigate to the Admin > Users tab. Under Invitations, enter the name or email ID of the person you want to invite and click Invite. This tab will show you a list of all outstanding invitations. Once an invitation has been accepted, the record will no longer show up on this page. The Users tab also displays a list of users of the application. Click on a username to see more details about the user.

If you want new users to join by invitation only, go to the Admin > Settings tab and check the Require invitation to sign up checkbox to require invitation tokens for account creation. By default, invitations are sent from noreply@your-cdsw-domain. To modify this default, see Setting up Email Notifications.

Assigning the Site Administrator role to an Existing User

To make a regular user a site administrator:
  1. Sign in to Cloudera Data Science Workbench as a site administrator.
  2. Click Admin.
  3. Click the Users tab.
  4. Click on the username of the user who you want to make a site administrator.
  5. Select the Site Administrator checkbox.
  6. Click Update.

Disabling User Accounts

Use the following instructions to disable user accounts. Note that disabled users cannot login and do not count towards named users for licensing.

  1. Sign in to Cloudera Data Science Workbench as a site administrator.
  2. Click Admin.
  3. Click the Users tab.
  4. Click on the username of the user who you want to disable.
  5. Select the Disabled checkbox.
  6. Click Update.

Monitoring Users

The Users tab on the admin dashboard displays the complete list of users. You can see which users are currently active, and when a user last logged in to the Cloudera Data Science Workbench. To modify a user's username, email or permissions, click the Edit button under the Action column.

Monitoring Site Usage

The Admin > Overview tab displays basic information about your deployment, such as the number of users signed up, the number of teams and projects created, memory used, and some average job scheduling and run times. You can also see the version of Cloudera Data Science Workbench you are currently running.

On the Admin > Activity tab of the dashboard, you can see the status for all the previous and current sessions running on the workbench.

Managing Engine Profiles

On the Admin > Engines page, under Engines Profiles, you can provide default engine profiles to users. These engine profiles provide default CPU and memory configurations for sessions and jobs. Cloudera recommends that all profiles include at least 2 GB of RAM to avoid out of memory errors for common user operations.

Adding Custom Engine Images

Cloudera Data Science Workbench site administrators can add libraries and other dependencies to the Docker image in which their engines run. You can whitelist images for use in projects on the Admin > Engines page, under the Engine Images section. Currently, Cloudera Data Science Workbench only supports public Docker images in registries accessible to the Cloudera Data Science Workbench nodes.

Project administrators will need to explicitly select which of these white-listed images is installed for their projects. For an example, see Customizing Engine Images.

Customizing the Engine Environment

On the Admin > Engines page, go to the Environmental Variables section to provide values for global environment variables that must be injected into all engines across the deployment.

Setting JAVA_HOME

By default, Cloudera Data Science Workbench will attempt to find your Java home. However, if you are using a non-standard path for Java, add JAVA_HOME under the Environmental Variables section. This will set JAVA_HOME and mount the JAVA_HOME directory into engines so that the Java version matches other CDH services.

Non-standard CDH Parcel Location

By default, Cloudera Data Science Workbench looks for the CDH parcel at /opt/cloudera/parcels. If your CDH parcel is at another location, add that path to the Parcel directory section.

Mounts

You can specify folders that should be mounted through to all the engines. Cloudera Data Science Workbench will automatically mount CDH parcels and client configuration.

Configuring External Authentication

Cloudera Data Science Workbench supports external authentication using LDAP and SAML protocols. You can configure LDAP or SAML authentication under the Admin > Security tab by following the instructions at Configuring External Authentication.

Setting up Email Notifications

Go to the Admin > Settings tab to specify an email address for outbound invitations and job notifications.

By default, all emails are sent from noreply@your-cdsw-domain. However, if your SMTP domain is different from the Cloudera Data Science Workbench domain, or it does not allow spoofing, you will need to explicitly specify the email address at the No Reply Email field.

Currently, Cloudera Data Science Workbench only sends email notifications when you add teammates to a project, not when you create a new project. Email preferences cannot currently be configured at an individual user level.

Managing License Keys

Cloudera Data Science Workbench is subject to the same license and subscription model as Cloudera Enterprise. To upload a license key, go to the Admin > License tab. For details on the types of licenses and detailed instructions for how to upload a new license key, see Managing License Keys for Cloudera Data Science Workbench.

Disabling Analytics Tracking

To help improve the product, Cloudera Data Science Workbench by default collects aggregate usage data by sending limited tracking events to Google Analytics and Cloudera servers. No customer data or personal information is sent as part of these bundles. To disable analytics tracking, go to Admin > Settings and uncheck the Send usage data to Cloudera checkbox.