Knowledge Platform

Edit | Report

Overview

This page explains the jobs to be run to bring up the Knowledge Platform services. In order to do so, log into Jenkins and execute the instructions as per the sequence provided on this page.

Build

Switch to the Build folder and run all jobs in the below sequence. For the value of the github_release_tag, refer to Current Release Tags and Jenkins Jobs Reference

Operation Name Function
CassandraTrigger Generate the jar file for cassandraTrigger
KnowledgePlatform Generate the artefacts for knowledge Platform
Neo4j Generate the Neo4j jar file
SyncTool Generate the SyncTool zip file
Yarn Generate the artefacts for Samza job

For KnowledgePlatform build, use the default profile_id which is “platform_services”.

DevOps Administration

Operation Name Function
Bootstrap Creates Deployer User

Provision

  • Download neo4j enterprise version 3.3.x. The file should be uploaded to your azure storage account under the root path. The file name should be neo4j*.tar.gz. This file should be uploaded to the container named in the variable of KnowledgePlatform’s common.yml.

The URL for this path will look like this - https://.blob.core.windows.net//neo4j-enterprise-3.3.0-unix.tar.gz

  • Switch to Provision/<env>/KnowledgePlatform and run the jobs in the following sequence:
Operation Name Function Description
Cassandra Provisions to install Cassandra Installs dependencies like java and installs Cassandra
CompositeSearch Installs the ElasticSearch Installs dependencies for elastic search and starts elastic search service
Neo4j Installs the Neo4j and Logstash Installs Neo4j and its dependencies
Zookeeper Installs the Zookeeper Zookeeper installation
Kafka Installs the Kafka Kafka installation
Learning Installs the dependency to run the learning application Creates learning user and installs dependencies(java, tomcat, logstash, ffmpeg,imagemagick)
Redis Installs the Redis database Download and installs Redis
Search Installs the dependency to run the search service and install the logstash Install dependencies (JDK, logstash) and configure search service
Yarn Sets up the Yarn cluster Setting up the yarn cluster

Deploy

  • Switch to Deploy/dev/KnowledgePlatform and run the jobs in the following sequence:
Operation Name Function
CassandraTrigger Deploys the CassandraTrigger jar file and installs the logstash
CassandraDbUpdate Creates the Cassandra keyspace and update
Neo4j Deploys the Neo4j artefacts
StartNeo4jCluster Starts the Neo4j process
Learning Deploys the learning service artefacts and starts the learning service
Search Deploys the search service artefacts and starts the search service
Neo4DefinitionUpdate Updates Neo4j definition
Neo4jElasticSearchSyncTool Deploys the sync tool artefacts and syncs the content from Neo4j to ElasticSearch
KafkaSetup Creates the Kafka Topics
Yarn Deploys the Samza jobs

Refer How to Create Framework, How to Create Schemas for Knowledge Platform Objects


Edit | Report