> osgi:install -s mvn:org.apache.servicemix.bundles/org.apache.servicemix.bundles.commons-httpclient/3.1_7
> features:install fabric-hadoop
* configure hadoop by creating a file in etc/org.fusesource.fabric.hadoop.cfg with the following content:
dfs.replication=1
mapred.job.tracker=localhost\:9001
dfs.name.dir=${karaf.data}/hadoop/dfs/name
dfs.http.address=localhost\:9002
dfs.data.dir=${karaf.data}/hadoop/dfs/data
dfs.name.edits.dir=${karaf.data}/hadoop/dfs/name
nameNode=true
dataNode=true
jobTracker=true
taskTracker=true
After those steps, you should have the hdfs web server available at
and the job tracker web server at
From this point, to deploy a job, you need to repackage your WordCount sample in an osgi bundle and modify it so that in a bundle activator (or using blueprint or any other similar technology), you create your job using the right configuration (the needed bits from the file above) and start it.
The hadoop support is still a bit work in progress and in particular, I have focused on hdfs rather than mapreduce, but I hope to simplify those steps asap.