# # Elastic MapReduce config in wukong # # =========================================================================== # # Infrastructure options # # == Fill all your information into yet another file with your amazon key # It needs to be in so many stupid places because nobody can agree on a # filename or format. # :emr_credentials_file: ~/.wukong/credentials.json # # == Set the AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY env vars, or enter them here # # :access_key: ASDFAHKHASDF # :secret_access_key: ADSGHASDFJASDFASDF # == Path to your keypair file. # :key_pair_file: ~/.wukong/keypairs/gibbon.pem # == Keypair will be named after your file, or force the name # # :key_pair: ~ # == Path to the Amazon elastic-mapreduce runner. Get a copy from # http://elasticmapreduce.s3.amazonaws.com/elastic-mapreduce-ruby.zip # :emr_runner: ~/ics/hadoop/elastic-mapreduce/elastic-mapreduce # =========================================================================== # # Remote Paths # # == Wukong is opinionated about the paths and locations of scripts and # everything. It will organize files by job name within the following path: # :emr_root: s3://s3n.infinitemonkeys.info/emr # == If you specify the :emr_data_root path, then relative pathnames -- ones that # do not look like a URI (s3://yadda/yada) and do not start with a '/' -- will # be prefixed with this path prefix. :emr_data_root: s3n://s3n.infinitemonkeys.info/data # =========================================================================== # # Cluster Config # :num_instances: 1 :instance_type: m1.small :master_instance_type: ~ :hadoop_version: '0.20' :availability_zone: us-east-1b # =========================================================================== # # Running and reporting options # :alive: true :enable_debugging: true :emr_runner_verbose: true :emr_runner_debug: ~ :step_action: CANCEL_AND_WAIT # CANCEL_AND_WAIT, TERMINATE_JOB_FLOW or CONTINUE