No such file or directory: slaves

[Neil Chaudhuri <Ne...@soteradefense.com>]

I have configured Whirr to start up some instances on EC2 as follows:

whirr.service-name=hadoop
whirr.cluster-name=graph-cluster
whirr.instance-templates=1 hadoop-jobtracker+hadoop-namenode, 3 hadoop-datanode+hadoop-tasktracker
whirr.provider=aws-ec2
whirr.location-id=us-east-1
whirr.hardware-id=m2.xlarge
whirr.cluster-user=${sys:user.name}

The instances start up correctly. Then I do the following

export HADOOP_CONF_DIR=~/.whirr/graph-cluster
$HADOOP_HOME/bin/start-all.sh

And I get this:

bigdata@clouddev:~/Desktop/whirr-0.8.1/bin? $HADOOP_HOME/bin/start-all.sh 255 ?
starting namenode, logging to /usr/lib/hadoop/logs/hadoop-bigdata-namenode-clouddev.out
cat: /home/bigdata/.whirr/graph-cluster/slaves: No such file or directory
cat: /home/bigdata/.whirr/graph-cluster/masters: No such file or directory
starting jobtracker, logging to /usr/lib/hadoop/logs/hadoop-bigdata-jobtracker-clouddev.out
cat: /home/bigdata/.whirr/graph-cluster/slaves: No such file or directory

I am sure I am doing something silly. Any help is appreciated.

Thanks.