#!/bin/sh
# Launch an EC2 cluster of Hadoop instances.

# Import variables
bin=`dirname "$0"`
bin=`cd "$bin"; pwd`
. "$bin"/hadoop-ec2-env.sh

ec2-describe-group | grep $GROUP > /dev/null
if [ ! $? -eq 0 ]; then
  echo "Creating group $GROUP"
  ec2-add-group $GROUP -d "Group for Hadoop clusters."
  ec2-authorize $GROUP -p 22    # ssh
  ec2-authorize $GROUP -p 50030 # JobTracker web interface
  ec2-authorize $GROUP -p 50060 # TaskTracker web interface
  ec2-authorize $GROUP -o $GROUP -u $AWS_ACCOUNT_ID 
fi

# Finding Hadoop image
AMI_IMAGE=`ec2-describe-images -a | grep $S3_BUCKET | grep $HADOOP_VERSION | grep available | awk '{print $2}'`

# Start a cluster
echo "Starting cluster with AMI $AMI_IMAGE"
RUN_INSTANCES_OUTPUT=`ec2-run-instances $AMI_IMAGE -n $NO_INSTANCES -g $GROUP -k $KEY_NAME -d "$NO_INSTANCES,$MASTER_HOST" | grep INSTANCE | awk '{print $2}'`
for instance in $RUN_INSTANCES_OUTPUT; do
  echo "Waiting for instance $instance to start"
  while true; do
    printf "."
    HOSTNAME=`ec2-describe-instances $instance | grep running | awk '{print $4}'`
    if [ ! -z $HOSTNAME ]; then
      echo "started as $HOSTNAME"
      break;
    fi
    sleep 1
  done
done

echo "Appointing master"
MASTER_EC2_HOST=`ec2-describe-instances | grep INSTANCE | grep running | awk '{if ($8 == 0 || $7 == 0) print $4}'`
MASTER_IP=`dig +short $MASTER_EC2_HOST`
echo "Master is $MASTER_EC2_HOST. Please set up DNS so $MASTER_HOST points to $MASTER_IP."