Solr Bootstrap is a simple set of scripts that will help you create a new solr cloud setup. Currently it only supports single node setups, but within that single node you can create as many collections/shards/replicas as you want.

solr-bootstrap/bin $ ./unpack_zookeeper.sh
solr-bootstrap/bin $ ./unpack_solr.sh
solr-bootstrap/bin $ ./start_zookeeper.sh
solr-bootstrap/bin $ ./start_solr.sh
solr-bootstrap/bin $ ./create_collection.sh sample 4 2
solr-bootstrap/bin $

Now you can navigate to http://localhost:12345/solr/#/~cloud to view your collection.

Use the create_collection.sh script to create a new collection. This assumes that there is a conf directory available at solr-bootstrap/data/solr/conf/. You can see the sample one for an example. The contents in that directory will be uploaded to zookeeper for configuration management. Note that after creating a collection updating that conf directory will have no impact on the collection. If you want to edit a conf then you will need to upload that into zookeeper. At the moment there is no script for that but you can always just delete the collection and recreate it.

solr-bootstrap/bin $ ./create_collection.sh <collectionName> <numberShards> <numberReplicas>

solr-bootstrap/bin $ ./delete_collection.sh <collectionName>

You can use index_data.sh to easily index data into your collection. This assumes you have a folder with all your data files in it.

solr-bootstrap/bin $ ./index_data.sh <collectionName> <dataDirectory> [data file extension, default='json']

You can edit anything in solr-bootstrap/bin/env.sh to change various settings (like the ports used by solr and zookeeper)