Architecture

PATROL for Hadoop enables you to monitor the Hadoop environment and resources. It also enables you to determine how to effectively use these resources. A typical architecture for PATROL for Hadoop is as follows:

Related topics

Key-concepts

Getting-started 


architecture_hadoop_v1.png

PATROL for Hadoop collector details

The following table lists different BMC PATROL for Hadoop collectors and explains how they work:

Collectors (Collection interval)

Explanation

ChannelListener
(2 seconds) 

Runs the listener method every two seconds to check the following:

  • If the Java collection should start for a new environment
  • If the Java collection should restart for a specific environment

Updates the MonitoringStatus parameter, if the Java collection fails to start for some reason

SetDataToPatrol
(1 second) 

Runs every second, to check if any data is received from Java. If yes, the parameter reads the data and sends it to PATROL Agent

ConfigurationManager
(15 seconds) 

Runs every 15 seconds to check if the Hadoop configurations are changed in the pconfig variables

AutoDiscovery
(23 hours 59 minutes 59 seconds)

Displays the time in seconds required to configure and discover the Hadoop environment on the Java side

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*