Merge pull request #7 from ntquach/SDAP-37

SDAP-37 Update SolrProxy to allow tile search by arbitrary tile metadata
tree: f9cbb341389c4b1974cbfea38ecf788308676229
  1. analysis/
  2. client/
  3. climatology/
  4. data-access/
  5. docker/
  6. esip-workshop/
  7. nexus-ingest/
  8. tools/
  9. CONTRIBUTING.md
  10. LICENSE
  11. README.md
README.md

nexus

The next generation cloud-based science data service platform

Developer Installation

  1. Follow instructions for installing nexusproto
  2. Follow instructions for installing data-access
  3. Follow instructions for setting up nexus-ingest
  4. Ingest some data
  5. Follow instructions for running analysis

Installation Instructions

Zookeeper Setup

  1. Install and configure Apache Zookeeper 3.4.x
  2. Create chroots solr, xd, and kafka

Database Setup

Apache Solr

  1. Install and configure Apache Solr Cloud 5.3.x

  2. Download JTS Topology Suite v1.13 and extract the zip.

  3. From the exploded JTS zip, copy $JTS_ZIP/lib/jts-1.13.jar and $JTS_ZIP/lib/jtsio-1.13.jar into $SOLR_INSTALL_DIR/server/lib/ext on all Solr nodes.

  4. Configure Solr Cloud to use the /solr chroot of zookeeper

  5. On one of the Solr cloud nodes, upload the nexustiles configuration (located in data-access/config/schemas/solr) as a configset

    ./zkcli.sh -cmd upconfig -z $ZK_SERVERS/solr -confname nexustiles -confdir /path/to/nexustiles/conf
    
  6. Create a new collection with the name nexustiles. Use the nexustiles configset uploaded previously.

    curl "http://<SOLR_HOST>/solr/admin/collections?action=CREATE&name=nexustiles&collection.configName=nexustiles"
    
  7. Repeat steps 5 & 6 for the datasets collection.

Apache Cassandra

  1. Install and configure Apache Cassandra 2.2.x
  2. Execute the DDL located in nexustiles.cql

HSQLDB

  1. Install and run HSQLDB 2.3.x

Redis

  1. Install and run Redis 3.0.x

Apache Kafka

  1. Install and configure Apache Kafka 2.11-0.9.0.1
  2. Configure Kafka to use the /kafka chroot of zookeeper

Spring XD

  1. Install Spring XD 1.3.1.RELEASE
  2. Optionally install Flo for Spring XD
  3. Configure to use previously installed Apache Kafka as messaging bus
  4. Configure to use previously installed Redis for analytics
  5. Configure to use previously installed HSQLDB for Job Repository
  6. Configure to use previously installed Apache Zookeeper using the xd chroot
  7. Choose and configure a location for the custom module registry
  8. Create a directory called none in $SPRING_XD_HOME/xd/lib

Preparing the XD Containers

In order for streams to use the custom python and groovy scripts found in nexus-ingest, the machines that will be running Spring XD containers need to have some software installed on them.

Anaconda

  1. Install Anaconda 4.0.0 with Python 2.7

  2. Create an Anaconda environment

    conda create --name nexus-xd-python-modules python
    
  3. Install conda dependencies

    conda install libnetcdf
    conda install netcdf4
    conda install numpy
    
  4. Install nexusproto

  5. Install spring-xd-python

  6. Install nexus-xd-python-modules

Java 8

  1. Install Java 8 JDK. Either Oracle or OpenJDK.

nexus-messages

  1. The nexus-messages JAR needs to be available via Maven. There are a number of ways to do this. One way is to explicitly install the dependency into the local maven cache on each XD Container node.

    mvn install:install-file -DpomFile=nexus-messages-VERSION.xml -Dfile=nexus-messages-VERSION.jar
    

Groovy Scripts

  1. Place the groovy-scripts in a location that will be accessible to all XD Container nodes

Custom Modules

  1. Upload the dataset-tiler custom module
  2. Upload the tcpshell custom module
  3. Upload the nexus-sink custom module