Dockerfile

# Dockerfile for the SC4 Pilot Floating Car Data producer 
#
# 1) Build an image using this docker file. Run the following docker command
#
#    $ docker build -t bde2020/pilot-sc4-fcd-producer:v0.1.0 .
#
# 2) Test the Kafka producer for the FCD data  in a container. Run the following docker command for testing
#
#    $ docker run --add-host=kafkahost:172.17.0.1 -it --name fcd-producer -e ZOOKEEPER_SERVERS=172.17.0.1 --rm bde2020/pilot-sc4-fcd-producer:v0.1.0 /bin/bash
#
#    The option --add-host tells docker to add kafkahost with its ipaddress to /etc/hosts file in the container so that the 
#    host name used in producer.props file in the bootstrap.servers=kafkahost:9092 can be resolved. 171.17.0.1 is the ipaddress
#    of the host computer. 
#    The environment variable ZOOKEEPER_SERVERS is used to wait for the kafka topic used by the producer to be available.
#    The Kafka broker, to which the producers send the data, must be configured in the server.properties file to listen to the
#    host network address assigned to it by docker, that is docker0 (not eth0). As an example if docker binds the network docker0
#    to the address 172.17.0.1 then in the server.properties file used to start a broker the listeners property must be set to
#    listeners=PLAINTEXT://172.17.0.1:9092   
#
# 3) Start a container with the  using the config file in the data volume
#
#    $ docker run -d --name fcd-producer bde2020/pilot-sc4-fcd-producer:v0.1.0 
#
#
#  We use the base from bde2020/kafka because all kafka command line tools and oracle java 8 are available there.
#  See: https://github.com/big-data-europe/docker-kafka

FROM bde2020/kafka
MAINTAINER Luigi Selmi <luigiselmi@gmail.com>, Karl-Heinz Sylla <karl-heinz.sylla@iais.fraunhofer.de>

# Put program/executable to application home
WORKDIR /home/pilot-sc4/

COPY target/pilot-sc4-kafka-producer-0.1.0-SNAPSHOT-jar-with-dependencies.jar .
COPY pilot-sc4-kafka-producer.sh .

# Run the FCD producer
CMD [ "./pilot-sc4-kafka-producer.sh" ]