The goals of this repo is to have an end to end working environment to provide the ability for systems (applications and infrastructure) to send their metrics/sensor data to Kafka and then to report on that data for both alerts and charts.
More about this project in this blog post http://allthingshadoop.com/2014/04/18/metrics-kafka/.
Use Vagrant to get up and running.
- Install Docker http://docs.docker.com/installation/
- Install Virtual Box https://www.virtualbox.org/
In the main metrics-kafka folder
- sudo ./bootstrap.sh
- ./gradlew test
- sudo ./shutdown.sh
once this is done
- Zookeeper will be running localhost:2181
- Broker 1 on localhost:9092
- Riemann on localhost:5555
- Riemann dash on localhost:4567
- All the tests in metrics/src/test/java/* and riemann/src/test/scala/* should pass
You can access the brokers, zookeeper, riemann and riemann-dash by their IP from your local without having to go into vm.
e.g.
bin/kafka-console-producer.sh --broker-list localhost:9092 --topic
bin/kafka-console-consumer.sh --zookeeper localhost:2181 --topic --from-beginning
http://192.168.86.5:4567/#Riemann ./gradlew :psutil:installDependencies ./gradlew :metrics-test:run
In order to assemble jar for metrics yammer do the following steps:
- ./gradlew :yammer:jar
- put the jar from metrics-yammer/build/libs to libs dir in root kafka folder
- add "kafka.metrics.reporters=ly.stealth.kafka.metrics.KafkaBrokerReporter" to config/server.properties in kafka root folder