Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with
or
.
Download ZIP
Shell
Branch: master

Fetching latest commit…

Cannot retrieve the latest commit at this time

Failed to load latest commit information.
bin
conf
.gitignore
README.md
install.sh

README.md

adevcloud

A fast way to get Accumulo up and running on your local machine. Well, it's not really that fast but hopefully it's a bit less painful than it can be.

It uses git submodules to track the necessary repos/branches from github:

  • accumulo (1.4.1)
  • hadoop-common (release-0.20.203.0)
  • zookeeper (release-3.3.6)

It then downloads all the source from github, compiles it, and fills in some config files for you.

How-To

Prerequisites

  • working Java installation
  • ant
  • maven2
  • password-less ssh setup

Get the repo, setup submodules

 git clone https://github.com/klucar/adevcloud.git
 cd adevcloud
 git submodule init
 git submodule update

now go get a cup of coffee or tea or redbull...

Compile everything

These are the commands in the buildall.sh script. You can run that and walk away for a while if you don't want to step through and wait for each build.

Hadoop

 cd hadoop-common
 ant binary
 tar xvf ./build/hadoop-0.20.203.1-SNAPSHOT-bin.tar.gz -C ../.
 cd ..

Zookeeper

 cd zookeeper
 ant binary
 tar xvf ./build/zookeeper-3.3.6-bin.tar.gz -C ../.
 cd ..

Accumulo

 cd accumulo
 mvn package && mvn assembly:single -N
 tar xvf target/accumulo-1.4.1-dist.tar.gz -C ../.
 cd ..

Set config files

The configall.sh script will prompt for directories where you want various files stored on the local filesystem. It then adds the appropriate values and copies config files where they need to go. Just pressing Enter will accept the default values.

 $ ./configall.sh 
      Hadoop Namenode Data Location : 
      ( /foo/bar/adevcloud/data/hadoop/nn ) > 
      Hadoop Datanode Data Location : 
      ( /foo/bar/adevcloud/data/hadoop/dn ) > 
      Map/Reduce Data Location : 
      ( /foo/bar/adevcloud/data/hadoop/mr ) > 
      Zookeeper Data Location : 
      ( /foo/bar/adevcloud/data/zookeeper ) > 

This script also sets the appropriate environment variables in your .bashrc file

Init Hadoop

From a new terminal

 $ /foo/bar/adevcloud/hadoop-0.20.203.1-SNAPSHOT/bin/hadoop namenode -format

Start Hadoop, Zookeeper

 $ /foo/bar/adevcloud/hadoop-0.20.203.1-SNAPSHOT/bin/start-all.sh
 $ /foo/bar/adevcloud/zookeeper-3.3.6/bin/zkServer.sh start

Init Accumulo

 $ /foo/bar/adevcloud/accumulo-1.4.1/bin/accumulo init

this will prompt for an instance name and password for the Accumulo root user

Start Accumulo

 $ /foo/bar/adevcloud/accumulo-1.4.1/bin/start-all.sh

Now go do something with Accumulo

$ /foo/bar/adevcloud/accumulo-1.4.1/bin/accumulo shell -u root

tl;dr

git clone https://github.com/klucar/adevcloud
cd adevcloud
git submodule init
git submodule update
./buildall.sh
./configall.sh
./initall.sh
./accumulo-1.4.1/bin/accumulo shell -u root
Something went wrong with that request. Please try again.