- If you need kafka (and Zookeeper) run the class KafkaLocal setting the input parameter to "true" (the local instance of Kafka will be executed with Zookeeper)
- If you need kafka and openTSDB then run the classes HbaseLocal and KafkaLocal setting the KafkaLocal's input parameter to "false"
Download avro-tools from http://mvnrepository.com/artifact/org.apache.avro/avro-tools/1.8.1 and exec:
java -jar path/to/avro-tools-1.8.1.jar compile schema src/main/resources/Event.avsc ./src/main/scala
To check the number of messagges in kafka using Claudera Manager:
ClauderaManager -> Kafka -> Chart Library -> Topics (click on topic to analyze)
To check the number of consumed messages in kafka using Claudera Manager:
ClauderaManager -> Yarn -> WebUI -> ResourceManager WebUi -> ApplicationMaster -> Streaming
Per generare il jar con tutte le dipendenze nella cartella ./lib
sbt universal:packageZipTarball Il tar si troverĂ nel path ./spark-opentsdb-examples/target/universal
Per aggiornare solo il jar senza dipendenze estrne
sbt clean package
spark-submit --executor-memory 1200M \
--jars $(JARS=("$(pwd)/lib"/*.jar); IFS=,; echo "${JARS[*]}") \
--driver-class-path /etc/hbase/conf \
--conf spark.executor.extraClassPath=/etc/hbase/conf \
--conf spark.executor.extraJavaOptions=-Djava.security.auth.login.config=/tmp/jaas.conf \
--conf "spark.driver.extraJavaOptions="-Dspark-opentsdb-exmaples.hbase.master=eligo105.eligotech.private:60000 -Dspark-opentsdb-exmaples.zookeeper.host=eligo105.eligotech.private:2181/kafka-Dspark-opentsdb-exmaples.kafka.brokers=192.168.2.108:9092" \
--master yarn --deploy-mode client \
--keytab flanotte.keytab \
--principal flanotte@SERVER.ELIGOTECH.COM \
--class com.cgnal.kafkaAvro.consumers.example.OpenTSDBConsumerMain spark-opentsdb-examples_2.10-1.0.0-SNAPSHOT.jar false flanotte.keytab flanotte@SERVER.ELIGOTECH.COM
A spark sbt template that you can use for bootstrapping your spark projetcs: http://www.davidgreco.me/blog/2015/04/11/a-spark-sbt-based-project-template/ http://blog.cloudera.com/blog/2015/03/how-to-tune-your-apache-spark-jobs-part-2/