dockerfiles
My dockerfiles inspired by many different implementations widely available on the internet. All images are based on java:8-jre image.
Currently available technologies:
- cassandra (with datastax agent)
- elasticsearch
- kafka
- kibana
- opscenter
- storm
- zookeeper
They might be run separately or as a whole cluster using docker-compose.
Usage
Build all images:
./build.sh
Start a cluster:
docker-compose up
Increase amount of cassandra, elasticsearch and storm-supervisor nodes to 3:
docker-compose scale scalable-cassandra=2 scalable-elasticsearch=2 storm-supervisor=3
Inject cassandra cluster information to opscenter:
./update-opscenter.sh
Direct access to container:
docker exec -i -t cassandra bash
Access to services
-
ElasticSearch
http://localhost:9200/ -
Kibana
http://localhost:5601/ -
OpsCenter
http://localhost:8888/ -
Storm UI
http://localhost:9999/ -
kafka-topics.sh --zookeeper $(docker inspect --format='{{ .NetworkSettings.IPAddress }}' zookeeper):2181 --create --replication-factor 1 --partition 1 --topic test -
kafka-topics.sh --zookeeper $(docker inspect --format='{{ .NetworkSettings.IPAddress }}' zookeeper):2181 --describe -
cqlsh $(docker inspect --format='{{ .NetworkSettings.IPAddress }}' cassandra) 9042 -
nodetool --host $(docker inspect --format='{{ .NetworkSettings.IPAddress }}' cassandra) status -
storm jar topology.jar com.your.package.AwesomeTopology -c nimbus.host=$(docker inspect --format='{{ .NetworkSettings.IPAddress }}' storm-nimbus) -c nimbus.thrift.port=6627