Ignorar navegação

Use Apache Kafka with Apache Storm on HDInsight

Última atualização: 07/05/2018
Editar no GitHub

This is a basic example of reading and writing string data to a Kafka on HDInsight cluster from Storm on HDInsight cluster.

NOTE: Apache Kafka and Storm are available as two different cluster types. HDInsight cluster types are tuned for the performance of a specific technology; in this case, Kafka and Storm. To use both together, you must create an Azure Virtual Network and then create both a Kafka and Storm cluster on the virtual network. For an example on how to do this using an Azure Resource Manager template, see the create-kafka-storm-clusters-in-vnet.json document. For an example of using the template with this example, see Use Apache Storm with Kafka on HDInsight.

Understanding the code

This project contains two topologys:

  • KafkaWriter: Defined by the writer.yaml file, this topology writes random sentences to Kafka using the KafkaBolt provided with Apache Storm.

    This topology uses a custom SentenceSpout component to generate random sentences.

  • KafkaReader: Defined by the reader.yaml file, this topology reads data from Kafka using the KafkaSpout provided with Apache Storm, then logs the data to stdout.

    This topology uses the HDFSBolt to log data read from Kafka to the HDFS compatible storage on the Storm cluster.


The topologies are defined using Flux. Flux was introduced in Storm 0.10.x (available on HDInsight version 3.3 and 3.4,) and allows you to separate the topology configuration from the code. For Topologies that use the Flux framework, the topology is defined in a YAML file. This can be included as part of the topology, or can be specified when you submit the topology to the Storm server. Flux also supports variable substitution at run-time, which is used in this example.

Both topologies use the dev.properties file at run time to provide the values for the following entries in the topologies:

  • ${kafka.topic}: The name of the Kafka topic that the topologies read/write to.

  • ${kafka.broker.hosts}: The hosts that the Kafka brokers run on. The broker information is used by the KafkaBolt when writing to Kafka.

  • ${kafka.zookeeper.hosts}: The hosts that Zookeeper runs on in the Kafka cluster.

To use

NOTE: These steps assume that you have a functioning Java development environment, including Maven 3.x

  1. Create an Azure Virtual Network that contains a Storm and Kafka cluster. Both clusters must be HDInsight 3.6.

  2. To enable the HDFSBolt to work with the HDFS compatible storage used by HDInsight, use the following script action on the Storm cluster:

    • Script URL: https://hdiconfigactions2.blob.core.windows.net/stormextlib/stormextlib.sh
    • Applies to: Nimbus and supervisor nodes
    • Persist: Yes

    For information on using the script, see the Customize HDInsight with script actions document.

  3. Download this repository

  4. From a command line in the project directory, use the following to build and package the topology.

    mvn clean package

    This will create a file named KafkaTopology-1.0-SNAPSHOT.jar in the target directory.

  5. Copy the KafkaTopology-1.0-SNAPSHOT.jar file to your Storm on HDInsight cluster. For example, scp KafkaTopology-1.0-SNAPSHOZT.jar myname@mystormcluster-ssh.azurehdinsight.net:

  6. Find the Zookeeper and Broker hosts for the Kafka cluster. The following examples show how to do this using PowerShell and Curl:

    • Brokers:

      $creds = Get-Credential -UserName "admin" -Message "Enter the HDInsight login"
      $clusterName = Read-Host -Prompt "Enter the Kafka cluster name"
      $resp = Invoke-WebRequest -Uri "https://$clusterName.azurehdinsight.net/api/v1/clusters/$clusterName/services/KAFKA/components/KAFKA_BROKER" `
          -Credential $creds
      $respObj = ConvertFrom-Json $resp.Content
      $brokerHosts = $respObj.host_components.HostRoles.host_name
      ($brokerHosts -join ":9092,") + ":9092"
      curl -su admin -G "https://$CLUSTERNAME.azurehdinsight.net/api/v1/clusters/$CLUSTERNAME/services/KAFKA/components/KAFKA_BROKER" | jq -r '["\(.host_components[].HostRoles.host_name):9092"] | join(",")'
    • Zookeeper:

      $creds = Get-Credential -UserName "admin" -Message "Enter the HDInsight login"
      $clusterName = Read-Host -Prompt "Enter the Kafka cluster name"
      $resp = Invoke-WebRequest -Uri "https://$clusterName.azurehdinsight.net/api/v1/clusters/$clusterName/services/ZOOKEEPER/components/ZOOKEEPER_SERVER" `
          -Credential $creds
      $respObj = ConvertFrom-Json $resp.Content
      $zookeeperHosts = $respObj.host_components.HostRoles.host_name
      ($zookeeperHosts -join ":2181,") + ":2181"
      curl -su admin -G "https://$CLUSTERNAME.azurehdinsight.net/api/v1/clusters/$CLUSTERNAME/services/ZOOKEEPER/components/ZOOKEEPER_SERVER" | jq -r '["\(.host_components[].HostRoles.host_name):2181"] | join(",")'
  7. Modify the dev.properties file to add the Zookeeper and Broker values from the previous step. Just copy and paste the string of comma-delimited fully qualified domain names that were returned.

    NOTE: The hdfs.url property in the dev.properties file is configured for a Storm cluster that uses an Azure Storage account. If your cluster uses Data Lake Store, change this value from wasb to adl.

  8. Upload the dev.properties file to the Storm cluster.

  9. Connect to the Kafka cluster using SSH. For example ssh sshuser@kafkaclustername-ssh.azurehdinsight.net.

  10. To create the topic used by the topology, use the following command. Replace $KAFKAZKHOSTS with the Zookeeper information retrieved earlier:

    /usr/hdp/current/kafka-broker/bin/kafka-topics.sh --create --replication-factor 3 --partitions 8 --topic stormtopic --zookeeper $KAFKAZKHOSTS
  11. Connect to Storm cluster using SSH. For example, ssh sshuser@stormclustername-ssh.azurehdinsight.net.

  12. Use the following command to start the writer topology:

    storm jar KafkaTopology-1.0-SNAPSHOT.jar org.apache.storm.flux.Flux --remote -R /writer.yaml --filter dev.properties

    The parameters used with this command are:

    • org.apache.storm.flux.Flux: Use Flux to configure and run this topology.
    • --remote: Submit the topology to Nimbus. This runs the topology in a distributed fashion using the worker nodes in the cluster.
    • -R /writer.yaml: Use the writer.yaml topology. -R indicates that this is a resource that is included in the jar file. It's in the root of the jar, so /writer.yaml is the path to it.
    • --filter dev.properties: Use the contents of dev.properties when starting the topology. This allows Flux to pick up the Kafka broker, Zookeeper, and topic values from the dev.properties. file. Flux uses these values in the reader.yaml file in place of the ${...} entries. For example, ${kafka.topic} is replaced by the value of kafka.topic: from the dev.properties file.
  13. Once the topology has started, use the following command from the SSH connection to view messages written to the stormtopic topic:

    NOTE: Replace $KAFKAZKHOSTS with the Zookeeper host information for the Kafka cluster.

     /usr/hdp/current/kafka-broker/bin/kafka-console-consumer.sh --zookeeper $KAFKAZKHOSTS --from-beginning --topic stormtopic

    This will start listing the random sentences that the topology is writing to the topic. Use Ctrl-c to stop the script.

  14. use the following command to start the reader topology:

    storm jar KafkaTopology-1.0-SNAPSHOT.jar org.apache.storm.flux.Flux --remote -R /reader.yaml --filter dev.properties
  15. Wait a minute and then use the following command to view the files created by the reader topology:

    hdfs dfs -ls /stormdata

    The output is similar to the following text:

    Found 173 items
    -rw-r--r--   1 storm supergroup       5137 2018-04-09 19:00 /stormdata/hdfs-bolt-4-0-1523300453088.txt
    -rw-r--r--   1 storm supergroup       5128 2018-04-09 19:00 /stormdata/hdfs-bolt-4-1-1523300453624.txt
    -rw-r--r--   1 storm supergroup       5131 2018-04-09 19:00 /stormdata/hdfs-bolt-4-10-1523300455170.txt
  16. Use the following command to stop the writer:

    storm stop kafka-writer
    storm stop kafka-reader