This work has been inspired by:
- ExoGENI Recipes: RENCI-NRIG/exogeni-recipes/accumulo
- Oracle Java 8: binarybabel/docker-jdk
- CentOS 7 base image: krallin/tini-images
Apache Accumulo is a key/value store based on the design of Google's BigTable. Accumulo stores its data in Apache Hadoop's HDFS and uses Apache Zookeeper for consensus. While many users interact directly with Accumulo, several open source projects use Accumulo as their underlying store.
See official documentation for more information.
$ docker build -t renci/accumulo:1.8.1 ./1.8.1/
...
$ docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
renci/accumulo 1.8.1 b392c045e54f 3 minutes ago 1.95GB
...
Automated builds are generated at: https://hub.docker.com/u/renci and can be pulled as follows.
$ docker pull renci/accumulo:1.8.1
Using the provided docker-compose.yml file to stand up a seven node Accumulo cluster that includes an accumulomaster, namenode, resourcemanager, two workers (worker1 and worker2) and two ZooKeeper nodes (zoo and zoo2).
Accumulo docker network and container port mapping (specific network values subject to change based on system):
The nodes will use the definitions found in the site-files directory to configure the cluster. These files can be modified as needed to configure your cluster as needed at runtime.
A docker volume named hadoop-public is also created to allow the nodes to exchange SSH key information between themselves on startup.
version: '3.1'
services:
accumulomaster:
image: renci/accumulo:1.8.1
depends_on:
- namenode
container_name: accumulomaster
volumes:
- hadoop-public:/home/hadoop/public
- ./site-files:/site-files
restart: always
hostname: accumulomaster
networks:
- accumulo
ports:
- '9995:9995'
environment:
ACCUMULO_MASTER: accumulomaster
IS_ACCUMULO_MASTER: 'true'
ACCUMULO_WORKERS: worker1 worker2
IS_ACCUMULO_WORKER: 'false'
IS_NODE_MANAGER: 'false'
IS_NAME_NODE: 'false'
IS_SECONDARY_NAME_NODE: 'false'
IS_DATA_NODE: 'false'
IS_RESOURCE_MANAGER: 'false'
CLUSTER_NODES: namenode resourcemanager worker1 worker2 accumulomaster
ZOOKEEPER_NODES: zoo1 zoo2
namenode:
image: renci/accumulo:1.8.1
depends_on:
- zoo1
container_name: namenode
volumes:
- hadoop-public:/home/hadoop/public
- ./site-files:/site-files
restart: always
hostname: namenode
networks:
- accumulo
ports:
- '50070:50070'
environment:
ACCUMULO_MASTER: accumulomaster
IS_ACCUMULO_MASTER: 'false'
ACCUMULO_WORKERS: worker1 worker2
IS_ACCUMULO_WORKER: 'false'
IS_NODE_MANAGER: 'false'
IS_NAME_NODE: 'true'
IS_SECONDARY_NAME_NODE: 'false'
IS_DATA_NODE: 'false'
IS_RESOURCE_MANAGER: 'false'
CLUSTER_NODES: namenode resourcemanager worker1 worker2 accumulomaster
ZOOKEEPER_NODES: zoo1 zoo2
resourcemanager:
image: renci/accumulo:1.8.1
depends_on:
- namenode
container_name: resourcemanager
volumes:
- hadoop-public:/home/hadoop/public
- ./site-files:/site-files
restart: always
hostname: resourcemanager
networks:
- accumulo
ports:
- '8088:8088'
environment:
ACCUMULO_MASTER: accumulomaster
IS_ACCUMULO_MASTER: 'false'
ACCUMULO_WORKERS: worker1 worker2
IS_ACCUMULO_WORKER: 'false'
IS_NODE_MANAGER: 'false'
IS_NAME_NODE: 'false'
IS_SECONDARY_NAME_NODE: 'false'
IS_DATA_NODE: 'false'
IS_RESOURCE_MANAGER: 'true'
CLUSTER_NODES: namenode resourcemanager worker1 worker2 accumulomaster
ZOOKEEPER_NODES: zoo1 zoo2
worker1:
image: renci/accumulo:1.8.1
depends_on:
- namenode
container_name: worker1
volumes:
- hadoop-public:/home/hadoop/public
- ./site-files:/site-files
restart: always
hostname: worker1
networks:
- accumulo
ports:
- '8042:8042'
- '50075:50075'
environment:
ACCUMULO_MASTER: accumulomaster
IS_ACCUMULO_MASTER: 'false'
ACCUMULO_WORKERS: worker1 worker2
IS_ACCUMULO_WORKER: 'true'
IS_NODE_MANAGER: 'true'
IS_NAME_NODE: 'false'
IS_SECONDARY_NAME_NODE: 'false'
IS_DATA_NODE: 'true'
IS_RESOURCE_MANAGER: 'false'
CLUSTER_NODES: namenode resourcemanager worker1 worker2 accumulomaster
ZOOKEEPER_NODES: zoo1 zoo2
worker2:
image: renci/accumulo:1.8.1
depends_on:
- namenode
container_name: worker2
volumes:
- hadoop-public:/home/hadoop/public
- ./site-files:/site-files
restart: always
hostname: worker2
networks:
- accumulo
ports:
- '8043:8042'
- '50076:50075'
environment:
ACCUMULO_MASTER: accumulomaster
IS_ACCUMULO_MASTER: 'false'
ACCUMULO_WORKERS: worker1 worker2
IS_ACCUMULO_WORKER: 'true'
IS_NODE_MANAGER: 'true'
IS_NAME_NODE: 'false'
IS_SECONDARY_NAME_NODE: 'false'
IS_DATA_NODE: 'true'
IS_RESOURCE_MANAGER: 'false'
CLUSTER_NODES: namenode resourcemanager worker1 worker2 accumulomaster
ZOOKEEPER_NODES: zoo1 zoo2
zoo1:
image: renci/zookeeper:3.4.11
container_name: zoo1
restart: always
hostname: zoo1
networks:
- accumulo
ports:
- 2181:2181
environment:
ZOO_MY_ID: 1
ZOO_SERVERS: server.1=0.0.0.0:2888:3888 server.2=zoo2:2888:3888
zoo2:
image: renci/zookeeper:3.4.11
container_name: zoo2
restart: always
hostname: zoo2
networks:
- accumulo
ports:
- 2182:2181
environment:
ZOO_MY_ID: 2
ZOO_SERVERS: server.1=zoo1:2888:3888 server.2=0.0.0.0:2888:3888
networks:
accumulo:
volumes:
hadoop-public:Using docker-compose
$ docker-compose up -d
After a few moments all containers will be running and should display in a ps call.
$ docker-compose ps
Name Command State Ports
-----------------------------------------------------------------------------------------------------------------
accumulomaster /usr/local/bin/tini -- /do ... Up 22/tcp, 0.0.0.0:9995->9995/tcp
namenode /usr/local/bin/tini -- /do ... Up 22/tcp, 0.0.0.0:50070->50070/tcp
resourcemanager /usr/local/bin/tini -- /do ... Up 22/tcp, 0.0.0.0:8042->8042/tcp, 0.0.0.0:8088->8088/tcp
worker1 /usr/local/bin/tini -- /do ... Up 22/tcp, 0.0.0.0:50075->50075/tcp
worker2 /usr/local/bin/tini -- /do ... Up 22/tcp, 0.0.0.0:50076->50075/tcp
zoo1 /usr/local/bin/tini -- /do ... Up 0.0.0.0:2181->2181/tcp, 2888/tcp, 3888/tcp
zoo2 /usr/local/bin/tini -- /do ... Up 0.0.0.0:2182->2181/tcp, 2888/tcp, 3888/tcp
Since the ports of the containers were mapped to the host the various web ui's can be observed using a local browser.
accumulomaster container: AccumuloMaster Web UI on port 9995
AccumuloMaster: http://localhost:9995/
namenode container: NameNode Web UI on port 50070
NameNode: http://localhost:50070/dfshealth.html#tab-datanode
resource manager container: ResourceManager Web UI on ports 8088
ResourceManager: http://localhost:8088/cluster
worker1 and worker2 containers: DataNode Web UI on ports 50075 and 50076, NodeManager on 8042 and 8043
DataNode (worker1): http://localhost:50075/datanode.html
NodeManager (worker1): http://localhost:8042/node
DataNode (worker2): http://localhost:50076/datanode.html
The cluster can be stopped by issuing a stop call.
$ docker-compose stop
Stopping worker2 ... done
Stopping accumulomaster ... done
Stopping worker1 ... done
Stopping resourcemanager ... done
Stopping namenode ... done
Stopping zoo1 ... done
Stopping zoo2 ... done
So long as the container definitions have not been removed, the cluster can be restarted by using a start call.
$ docker-compose start
Starting zoo1 ... done
Starting zoo2 ... done
Starting namenode ... done
Starting worker2 ... done
Starting worker1 ... done
Starting resourcemanager ... done
Starting accumulomaster ... done
After a few moments all cluster activity should be back to normal.
The entire cluster can be removed by first stopping it, and then removing the containers from the local machine.
$ docker-compose stop && docker-compose rm -f
Stopping worker1 ... done
Stopping resourcemanager ... done
Stopping accumulomaster ... done
Stopping worker2 ... done
Stopping namenode ... done
Stopping zoo1 ... done
Stopping zoo2 ... done
Going to remove worker1, resourcemanager, accumulomaster, worker2, namenode, zoo1, zoo2
Removing worker1 ... done
Removing resourcemanager ... done
Removing accumulomaster ... done
Removing worker2 ... done
Removing namenode ... done
Removing zoo1 ... done
Removing zoo2 ... done
NOTE: Assumes the cluster is running as configured in the previous example.
A script named usertable-example.sh will create a sample usertable in Accumulo using 100 randomly generated user entries. Calls to the accumulomaster container are made using docker exec.
The user can also invoke the accumulo shell with the following command.
$ docker exec -ti accumulomaster runuser -l hadoop -c '${ACCUMULO_HOME}/bin/accumulo shell -u root -p secret'
Running usertable-example.sh:
$ ./usertable-example.sh
INFO: generate splits.txt
user2630
user6754
...
user1279
user2634
docker cp splits.txt accumulomaster:/tmp/splits.txt
INFO: ${ACCUMULO_HOME}/bin/accumulo shell -u root -p secret -e "deletetable -f usertable"
2018-02-19 15:35:10,017 [trace.DistributedTrace] INFO : SpanReceiver org.apache.accumulo.tracer.ZooTraceClient was loaded successfully.
2018-02-19 15:35:10,211 [shell.Shell] ERROR: org.apache.accumulo.core.client.TableNotFoundException: Table usertable does not exist
INFO: ${ACCUMULO_HOME}/bin/accumulo shell -u root -p secret -e "createtable usertable"
2018-02-19 15:35:13,920 [trace.DistributedTrace] INFO : SpanReceiver org.apache.accumulo.tracer.ZooTraceClient was loaded successfully.
INFO: ${ACCUMULO_HOME}/bin/accumulo shell -u root -p secret -e "addsplits -t usertable -sf /tmp/splits.txt"
2018-02-19 15:35:17,697 [trace.DistributedTrace] INFO : SpanReceiver org.apache.accumulo.tracer.ZooTraceClient was loaded successfully.
INFO: ${ACCUMULO_HOME}/bin/accumulo shell -u root -p secret -e "config -t usertable -s table.cache.block.enable=true"
2018-02-19 15:35:25,261 [trace.DistributedTrace] INFO : SpanReceiver org.apache.accumulo.tracer.ZooTraceClient was loaded successfully.
INFO: ${ACCUMULO_HOME}/bin/accumulo shell -u root -p secret -e tables
2018-02-19 15:35:28,434 [trace.DistributedTrace] INFO : SpanReceiver org.apache.accumulo.tracer.ZooTraceClient was loaded successfully.
accumulo.metadata
accumulo.replication
accumulo.root
trace
usertable
AccumuloMaster: http://localhost:9995/master
- ExoGENI Accumulo: https://github.com/RENCI-NRIG/exogeni-recipes/blob/master/accumulo/accumulo_exogeni_postboot.sh
- Accumulo Docs: https://accumulo.apache.org/1.8/accumulo_user_manual.html
- Hadoop Docs:
- Common: hadoop-common/core-default.xml
- HDFS: hadoop-hdfs/hdfs-default.xml
- MapReduce: hadoop-mapreduce-client-core/mapred-default.xml
- Yarn: hadoop-yarn-common/yarn-default.xml
- Deprecated Properties: hadoop-common/DeprecatedProperties.html
- ZooKeeper Docs: https://zookeeper.apache.org/doc/r3.4.11/zookeeperAdmin.html