Getting Started

Lets start with setting up Kafkawize on your Windows/Linux/Unix systems.

Prerequisites

  • Any IDE like IntelliJ/Netbeans or Eclipse to build the applications.
  • Java installed machines to run the applications
  • Rdbms (like Oracle/MySQL..) or Apache Cassandra

1 Download Kafkawize

Download the latest version (3.3) or clone from the git repo. https://github.com/muralibasani/kafkawize

Download the latest version (3.2) or clone from the git repo. https://github.com/muralibasani/kafkawizeclusterapi

OR

Download Kafkawize 3.3 bundle here Kafkawize_3.3

Previous version 3.2 https://kafkawize.readthedocs.io/en/stable/getting_started.html

2 Download Metastore

If Cassandra is used as metastore, you can download from http://cassandra.apache.org/download/

If Rdbms is used as metastore, download

Mysql : https://dev.mysql.com/downloads/

Oracle : https://www.oracle.com/database/technologies/oracle-database-software-downloads.html

Or any other Rdbms is ok.

3 Build KW Cluster Api Application

This Api does the below

  • Receives and responds to calls from UI Api
  • Connects to Kafka Brokers with Kafka AdminClient Api
  • There is no connection to any metastore
  • Swagger enabled

Unzip kafkawizeclusterapi.zip and open in any IDE

Configure the application properties (src/main/resources) if port has to be changed, else default port is 9343

and run maven command to create a runnable jar:

mvn clean package

This should create a jar (kafkawizeclusterapi-3.2.jar) in target dir.

Kafka connectivity between ClusterApi application and Kafka cluster, if SSL connection needs to be configured, configure “environment”.connect_with_ssl_kafkacluster in application properties to true and configure the other keystore properties even.

Run:

java -jar kafkawizeclusterapi-3.2.jar

4 Metastore setup

Metastore Cassandra

  • Install and run Cassandra and create a keyspace ‘kafkamanagementapi’

  • Create tables and run insert scripts in Cassandra

    /kafkawize/kafkawize-web/src/main/resources/scripts/base/cassandra/createcassandra.sql

    /kafkawize/kafkawize-web/src/main/resources/scripts/base/cassandra/insertdata.sql

    (Scripts available in kafkawize.zip)

  • Above scripts will create tables and insert initial set of Environments, Teams and Users which you can delete anytime from UI.

Metastore Rdbms

  • Install and run Mysql/Oracle and create a db schema or database

  • Create tables and run insert scripts in Database

    /kafkawize/kafkawize-web/src/main/resources/scripts/base/rdbms/ddl-jdbc.sql

    /kafkawize/kafkawize-web/src/main/resources/scripts/base/rdbms/insertdata.sql

    (Scripts available in kafkawize.zip)

  • Above scripts will create tables and insert initial set of Environments, Teams and Users which you can delete anytime from UI.

5 Build KW UI Api Application

This Api does the below

  • Users interact with interface with this Api
  • All the end points in this application either connect to Metastore or Cluster Api or both

Unzip kafkawize.zip and open in any IDE

Configure application properties:

In kafkawize/src/main/resources/application.properties

default port is 9097, if port has to be changed, else

server.port:9097

If metastore is cassandra (from step 4)

setstore type as cassandra:

db.storetype=cassandra

configure cassandra host, port and keyspace:

cassandradb.keyspace:kafkamanagementapi
cassandradb.url:localhost
cassandradb.port:9042

If metastore is rdbms (from step 4)

setstore type as rdbms:

db.storetype=rdbms
  • Install and run Rdbms (like Mysql/Oracle) and create a db schema or database

configure db properties like below:

# Spring JPA properties
spring.datasource.url=jdbc:mysql://localhost:3306/kafkametadb?autoReconnect=true&useUnicode=true&useJDBCCompliantTimezoneShift=true&useLegacyDatetimeCode=false&serverTimezone=UTC
spring.datasource.username=kafkauser
spring.datasource.password=kafkauser123
spring.datasource.driver.class=com.mysql.cj.jdbc.Driver

Configure Cluster Api

configure cluster api host and port details:

clusterapi.url:http://localhost:9343
  • ignore user/pwd of cluster api properties

Build

Run maven command to create a runnable jar:

mvn clean package

This should create a jar in target dir (/kafkawize/target/kafkawize-3.3.jar).

Run:

java -jar spring.config.location=classpath:/application.properties kafkawize-3.3.jar

If application is running, you can access UI from http://[host]:[port]/kafkawize

6 Kafka Connectivity

Cluster Api Application connects to Kafka brokers with Kafka AdminClient Api., and needs Describe access on all topics through the cluster. Hence the below wildcard acl has to be executed.

  • If Acls are enabled on Kafka brokers, make sure “Cluster Api” application host is authorized to read topics (A read Acl is enough on the topic)

    Examples SSL Based Acl:

    bin/kafka-acls –authorizer-properties zookeeper.connect=localhost:2181 –add –allow-principal User:”CN=MO,OU=MO,O=WA,L=WA,ST=WA,C=HO” –operation All –topic “*” –cluster

    Examples IP Based Acl:

    bin/kafka-acls –authorizer-properties zookeeper.connect=localhost:2181 –add –allow-principal User:”*” –allow-host 127.0.0.1 –operation All –topic “*” –cluster

  • If SASL/SSL is configured, make sure they right properties are configured in AdminClient properties in Cluster Api application.

7 Final Check