About the Cloudera Observability On-Premises user interface hierarchy

Provides a brief introduction to the web interface, its hierarchical components, and the frequently used interface elements of Cloudera Observability On-Premises. Cloudera recommends that you take a moment to familiarize yourself with the user interface, its components, and elements.

The Cloudera Observability On-Premises web UI hierarchically displays the health, performance, and status of your environment, workload clusters, engines, and resources, including the costs associated with your data infrastructure, from the top-down. Its dashboard components include statistics, performance, health, and prescriptive guidance visually displayed in chart widgets, cards, and tabular views.

After logging in to the Cloudera Observability On-Premises UI the landing page opens. This page is also called the Cloudera Observability On-Premises HOME page.

About the main navigation panel

The Main navigation panel enables access to the following Cloudera Observability On-Premises features:
  • Financial Governance, which opens the Chargeback page that displays the total costs and the hourly CPU and memory usage for all of your cost centers, including the unutilized resource usage costs from the Uncategorised section.
    From its Actions list you can:
    • Configure your cost centers criteria based on CPU and Memory costs and resource usage.
    • Create cost centers, which separate costs across user or pool usage and track their workload resource consumption costs.
  • Analytics, which depending on the tier level within the hierarchy displays:
    • The Environments page that lists your environments, including:
      • The environment's platform type. At this time, only Private Cloud Base environments are supported
      • The platform's version number.
      • The type of data content.
      • The date and time that telemetry data was last collected.

      Where, a Cloudera Observability On-Premises Environment hierarchically represents the association of your environment and its resources, clusters, engines, and their workloads (Jobs and Queries).

      To display clusters that have not sent data in the last 30 days, select the Show stale clusters check box.

  • Access Management, which opens the Access Management page for managing your Cloudera Observability On-Premises cluster policies and access roles when the Role Based Access property is enabled in Cloudera Manager.
    A Cloudera Observability On-Premises Cluster Policy defines the conditions for a role based access type. For example, they define:
    • Who is entitled to access jobs and queries that are created by the user.
    • Who is entitled to create and administer cost centers and view cluster costs.
    • Who is entitled to access and administer jobs and queries within either a specific cluster or across all clusters within the Cloudera Observability On-Premises environment.

    Limiting the trust boundary for jobs, queries, cluster costs, and administrative management at the cluster level, enables more control over the security and access management of your Cloudera Observability On-Premises environment.

  • User login name, which enables you to securely log out.

About the environment navigation panel

The Environments navigation panel hierarchically displays the environment from its parent tier (environment name) to the lower tier levels (clusters, engines, jobs and queries, and, where applicable, your Hive Metastore).

The following table describes the environment categories, which are displayed in the Environment panel dependent on the selected Cloudera Observability On-Premises environment type:
Table 1. Environment categories
Category Description
ENGINES Lists the Hive, Impala, Spark, Oozie, and MapReduce workload engines for Private Cloud Base environments.
HIVE METASTORE Lists the Hive and Impala engine metastores.

When a metastore is selected the HMS Summary view opens displaying information about the current state and activity of all your tables in the selected Environment. To display details about each table in your system that were processed in Hive and Impala engines, regardless of whether they have been queried or not, select the Tables tab, which opens the HMS Tables view.

The following table describes the Cluster, and Engine Summary pages:
Table 2. Environment summary pages
Cluster Summary Displays the Summary page and the Cloudera Observability On-Premises features available for the environment's cluster and subscription entitlement:
  • Summary tab, which displays performance trends and metrics about the processed jobs and queries as chart widgets and enables you to view historical trends for analysis when you select a predefined or custom time period from the Time-Range filter list.
  • Workloads tab, which enables the creation and management of your Workload views for the environment's cluster.
  • Auto Actions tab, which enables the creation and management of your Auto Actions events for the environment's cluster.
Engine Summary Displays information about the workload jobs or queries run by the selected engine, such as which jobs or queries have failed or are slow, their processing time, missed SLAs (thresholds), user and pool metrics, and outlier issues.

About the Cloudera Observability On-Premises chart widgets

The Cloudera Observability On-Premises chart widgets enable you to quickly observe real-time and historical patterns, trends, and outliers of your workload data. They provide quick insights into the health and performance of your workloads, clusters, and resources. Where:
  • Hovering over an element with your mouse pointer, such as over a time-line or a data point, displays more information about the element underneath.
  • Clicking a link within a chart widget or a bar within bar chart widget, such as the Suboptimal bar chart widget, opens the engine's Jobs or Queries page that contains more information in a tabular view for you to investigate further.

About the Jobs and Queries pages

Depending on the engine chosen, the engine's Jobs or Queries page provides information about each of the jobs or queries that are serviced by the engine. You can filter further by selecting a filter option from one of the filter categories; Pool, User, Status, Health Check, or Duration.

For more information about a specific job's or query's health, execution details, baseline, and trends, open their respective page by clicking the link in the Job or Query column and selecting the tab of interest. To investigate further, these pages also provide prescriptive guidance and recommendations that enable you to address problems and optimize solutions.