Cloudera Docs

Monitoring your Hive service

Identify Hive service problems that may be affecting your Hive workloads, such as queries that are running slow or that are failing, with the Cloudera Observability On-Premises Hive cluster service metrics.

Describes where to view the Cloudera Observability On-Premises Hive cluster service metrics and how to build your own service chart from the Cloudera Manager service metrics and Chart Builder.
  1. Verify that you are logged in to the Cloudera Observability On-Premises web UI.
    1. In the URL field of a supported web browser, enter the Cloudera Observability On-Premises URL that you were given by your system administrator and press Enter.
    2. When the Cloudera Observability On-Premises Log in page opens, enter your Cloudera Observability On-Premises user name and password access credentials.
    3. Click Log in.
      The Cloudera Observability On-Premises web user interface landing page opens, which by default displays the Analytics Environments page that lists your Workload cluster environments.
    4. From the Environment Name column in the Environment's table, select the environment required for analysis.

      The Environments navigation panel opens, which hierarchically lists the environment's cluster, engines, and if applicable the Hive Metastore category.

  2. Depending on the environment selected, verify that the Cluster Summary page is displayed for the environment's cluster required for analysis.
  3. Optional: From the time-range list in the Cluster Summary page, select a time period that meets your requirements.
  4. Locate the Hive query of interest by doing one of the following:
    • In the Cluster Summary page, locate the Hive Query Trend chart widget, click its Total Queries value, and then from the Job column in the Queries page, locate and click the query of interest.
    • If not already expanded, from the Environment navigation panel, expand the ENGINES category and then select the Hive engine. Locate the Slow Queries chart widget and click the query of interest.
  5. From the .../Hive/Queries/ queryname page, select the Cluster tab.
    The Analysis page opens, which lists the Hive cluster health check metrics that are performed by Cloudera Observability On-Premises at the end of a Hive job.
  6. Select, either the metric you require for analysis or a metric that displays a red or yellow icon adjacent to the metric, which represents the threshold warning or error state for at least one unhealthy role instance.
    The Analysis summary page opens, which describes the health check metric performed by Cloudera Observability On-Premises on the Hive service, the severity conditions and thresholds, and the remediation actions you should consider to resolve a problem. It also contains:
    • The Analysis chart, which displays the severity condition of the operation during the job run and displays the state of each role instance 5 minutes before the start of the job and 5 minutes after the job has completed.
    • The Host Status section, which displays the full list of workload role instances and the hosts they are running on, their health check result, and the severity state icon.
  7. Optional: To build your own chart from a Cloudera Manager health check service metric, click the Metrics option and do the following:
    1. From the Service Name list, select a service that you are running on your workloads cluster.
    2. From the Metric Name list, select the name of the Cloudera Manager health test metric.
    3. Click View.