Monitoring a Cluster Using Cloudera Manager

Cloudera Manager provides many features for monitoring the health and performance of the components of your clusters (hosts, service daemons) as well as the performance and resource demands of the jobs running on your clusters.

The following monitoring features are available in Cloudera Manager:

  • Monitoring Cloudera Runtime Services - describes how to view the results of health tests at both the service and role instance level. Various types of metrics are displayed in charts that help with problem diagnosis. Health tests include advice about actions you can take if the health of a component becomes concerning or bad. You can also view the history of actions performed on a service or role, and can view an audit log of configuration changes.
  • Monitoring Hosts - describes how to view information pertaining to all the hosts on your cluster: which hosts are up or down, current resident and virtual memory consumption for a host, what role instances are running on a host, which hosts are assigned to different racks, and so on. You can look at a summary view for all hosts in your cluster or drill down for extensive details about an individual host, including charts that provide a visual overview of key metrics on your host.
  • Activities - describes how to view the activities running on the cluster, both at the current time and through dashboards that show historical activity, and provides many statistics, both in tabular displays and charts, about the resources used by individual jobs. You can compare the performance of similar jobs and view the performance of individual task attempts across a job to help diagnose behavior or performance problems.
  • Events - describes how to view events and make them available for alerting and for searching, giving you a view into the history of all relevant events that occur cluster-wide. You can filter events by time range, service, host, keyword, and so on.
  • Alerts - describes how to configure Cloudera Manager to generate alerts from certain events. You can configure thresholds for certain types of events, enable and disable them, and configure alert notifications by email or using SNMP trap for critical events. You can also suppress alerts temporarily for individual roles, services, hosts, or even the entire cluster to allow system maintenance/troubleshooting without generating excessive alert traffic.
  • Lifecycle and Security Auditing - describes how to view service, role, and host lifecycle events such as creating a role or service, making configuration revisions for a role or service, decommissioning and recommissioning hosts, and running commands recorded by Cloudera Manager management services. You can filter audit event entries by time range, service, host, keyword, and so on.
  • Charting Time-Series Data - describes how to search metric data, create charts of the data, group (facet) the data, and save those charts to user-defined dashboards.
  • Logs - describes how to access logs in a variety of ways that take into account the current context you are viewing. For example, when monitoring a service, you can easily click a single link to view the log entries related to that specific service, through the same user interface. When viewing information about a user's activity, you can easily view the relevant log entries that occurred on the hosts used by the job while the job was running.
  • Reports - describes how to view historical information about disk utilization by user, user group, and by directory and view cluster job activity user, group, or job ID. These reports are aggregated over selected time periods (hourly, daily, weekly, and so on) and can be exported as XLS or CSV files. You can also manage HDFS directories as well, including searching and setting quotas.
  • Troubleshooting Cluster Configuration and Operation - contains solutions to some common problems that prevent you from using Cloudera Manager and describes how to use Cloudera Manager log and notification management tools to diagnose problems.