Dear Wiki user, You have subscribed to a wiki page or wiki category on "Hadoop Wiki" for change notification.
The following page has been changed by JimRWilson: http://wiki.apache.org/hadoop/AmazonEC2 The comment on the change is: fixing old, broken svn download links ------------------------------------------------------------------------------ Here is a high-level outline of how the scripts for creating a Hadoop AMI work. For details, please see the scripts' sources (linked to below). - 1. The main script, [http://svn.apache.org/viewvc/lucene/hadoop/trunk/src/contrib/ec2/bin/create-hadoop-image?view=co create-hadoop-image] starts a Fedora core Amazon AMI. + 1. The main script, [http://svn.apache.org/viewvc/hadoop/core/trunk/src/contrib/ec2/bin/create-hadoop-image?view=co create-hadoop-image] starts a Fedora core Amazon AMI. - 1. Once the Fedora instance has launched ''create-hadoop-image'' copies the environment variables file ([http://svn.apache.org/viewvc/lucene/hadoop/trunk/src/contrib/ec2/bin/hadoop-ec2-env.sh.template?view=co hadoop-ec2-env.sh]) and scripts to run on the Fedora instance ([http://svn.apache.org/viewvc/lucene/hadoop/trunk/src/contrib/ec2/bin/image/create-hadoop-image-remote?view=co create-hadoop-image-remote] and [http://svn.apache.org/viewvc/lucene/hadoop/trunk/src/contrib/ec2/bin/image/hadoop-init?view=co hadoop-init]) then it logs into the Fedora instance and runs ''create-hadoop-image-remote''. + 1. Once the Fedora instance has launched ''create-hadoop-image'' copies the environment variables file ([http://svn.apache.org/viewvc/lucene/hadoop/trunk/src/contrib/ec2/bin/hadoop-ec2-env.sh.template?view=co hadoop-ec2-env.sh]) and scripts to run on the Fedora instance ([http://svn.apache.org/viewvc/hadoop/core/trunk/src/contrib/ec2/bin/image/create-hadoop-image-remote?view=co create-hadoop-image-remote] and [http://svn.apache.org/viewvc/hadoop/core/trunk/src/contrib/ec2/bin/image/hadoop-init?view=co hadoop-init]) then it logs into the Fedora instance and runs ''create-hadoop-image-remote''. 1. The script ''create-hadoop-image-remote'' then installs Java, tools required to run Hadoop, and Hadoop itself. Then it configures Hadoop: * In EC2, the local data volume is mounted as ''/mnt'', so logs are written under here. * ''hadoop-init'' is installed as an init script to be run on instance start up. This takes advantage of an EC2 feature called [http://developer.amazonwebservices.com/connect/entry.jspa?externalID=531&categoryID=100 parameterized launches]. For Hadoop, this allows the master hostname and the cluster size to be retrieved when the Hadoop instance starts - this information is used to finish the Hadoop configuration.
