Kill

Storm topologies are never-ending processes. To stop a topology, we need to kill it. When killed, the topology first enters into the deactivation state, processes all the tuples already emitted into it, and then stops. Run the following command to kill SampleStormClusterTopology:

$> bin/storm kill SampleStormClusterTopology  

The following information is displayed:

0 [main] INFO backtype.storm.thrift - Connecting to Nimbus at localhost:6627 
80 [main] INFO backtype.storm.command.kill-topology - Killed topology: SampleStormClusterTopology
  

Now, run the jps command again to see the remaining JVM processes as follows:

jps  

The preceding command's output is:

26530 supervisor 
27193 Jps 
26468 nimbus