#Instructions: how to use In this file you can find instructions how to use CEP-esper-consumer project as well as project from repository (masters-thesis-kafka-producer)

##List of the machines:

• huzvak01 - Node with Kafka broker and ZooKeeper
• huzvak02 - Node aimed for Esper project, also contains ZooKeeper
• huzvak03 - Node aimed for Esper project, also contains ZooKeeper
• huzvak04 - Node aimed for Esper project, also contains ZooKeeper

##How to connect to the machines:

• you have to have your public ssh key on the server (currently only author, supervisor and oponent)
• connect through following commands from command line:
• ssh huzvak-01@147.251.43.240
• ssh huzvak-02@147.251.43.218
• ssh huzvak-03@147.251.43.236
• ssh huzvak-04@147.251.43.219

##How to run Kafka broker:

• you have to be connected to huzvak01 machine
• move to diplomka/kafka/kafka_2.9.2-0.8.1.1/ folder located in home folder (of user huzvak-0x) with following command:
• cd diplomka/kafka/kafka_2.9.2-0.8.1.1/
• run Kafka broker with following command:
• ./bin/kafka-server-start.sh ./config/server.properties
• to stop Kafka broker just press Control + C

##How to run ZooKeeper command line interface (ZK cli):

• you have to be connected to any from above mentioned four machines (probably it will works from any machine that has installed ZooKeeper and is in the Masaryk University network)
• move to folder /usr/share/zookeeper/bin/ located in home folder (of user huzvak-0x) with following command:
• cd /usr/share/zookeeper/bin/
• run ZK cli with following command:
• ./zkCli.sh -server 147.251.43.240:2181,147.251.43.218:2181,147.251.43.236:2181,147.251.43.219:2181

##Commands for ZK cli:

• for all commands just type help and press enter
• here are the most used commands:
• create /path/to/znode data - this command create new node on given path and store given data into it
• get /path/to/znode - return all the information together with data of the znode on given path
• ls /path/to/znode - list all children of the znode
• set /path/to/znode data dataversion - set the data into given znode. Valid only if correct dataversion is provided, otherwise fails. The dataversion number can be read from output from get command
• delete /path/to/znode dataversion - delete znode on given path if correct dataversion is provided

##How to run (Producer)

• (can be run from any computer connected to the MU network, and with at least Java 1.7 installed)
• move to the folder where the jar file with dependencies is located
• run following command (the number at the end, is the number of topic to which data should be produced):
• java -jar kafka-producer-1.0-SNAPSHOT-jar-with-dependencies.jar 1

##How to run Esper and consumer (cep-esper-consumer project)

• (should be run from machines huzvak02, huzvak03 and huzvak04)
• currently cannot be run from other machines in MU network, because ZK timeout exception is set to 400 ms which is too low
• when connected to the machine, move to folder projekt in home folder (of user huzvak-0x). You can use following command:
• cd ./projekt
• to start consumer, use the following command, where arguments are following:
• 1st argument is the znode from which configuration should be read
• 2nd is the consumer group id
• 3rd is the string with IP addresses and ports to ZooKeeper nodes
• example of the command is:
• java -jar CEP-esper-consumer-0.0.1-SNAPSHOT-jar-with-dependencies.jar /esper/node1 1 147.251.43.240:2181,147.251.43.218:2181,147.251.43.236:2181,147.251.43.219:2181