Cloudera Manager | Hadoop Service Monitoring Demo Video
In this demo video, BC Wong, a software engineer at Cloudera, discusses the Hadoop Service Monitoring feature in Cloudera Manager. Service Monitoring helps you monitor and manage your Hadoop clusters effectively.
Through the Service Monitoring feature, customers can monitor dozens of service health and performance metrics about the overall service (HDFS, MapReduce, HBase). They can also examine underlying role instances (Namenode, Datanodes, JobTracker, TaskTrackers, Region Servers etc.) in your Hadoop cluster and see what’s going wrong – or what is about to go wrong.
Service Monitoring presents health and performance data in a variety of formats including interactive charts through Cloudera’s new, enhanced user interface. Every Service Monitoring page also includes a widget to enable quick search for relevant Events and Logs associated with the service under consideration. Important Event and Log messages are also highlighted in the various charts. You can also monitor metrics against customizable thresholds, which results in Alerts that operators can pay attention to.
This demo covers examples of identifying problems with a service, underlying roles and instances. It reviews the service summary and health checks, logs and events to determine the root cause of the problem.