Bigdata Dev Apache Hadoop Resourcemanager

Learn about actions >

• downgrade

  Downgrade this unit

• finalize

  Finalize and commit an in-progress upgrade

• list-versions

  list available versions of hadoop

• mrbench

  Mapreduce benchmark for small jobs

  Params

  • basedir string

    DFS working directory

  • inputlines integer

    number of input lines to generate

  • inputtype string

    Type of input to generate, one of [ascending, descending, random]

  • maps integer

    number of maps for each run

  • numruns integer

    Number of times to run the job

  • reduces integer

    number of reduces for each run

• prepare-upgrade

  Prepare HDFS for an upgrade

• restart-yarn

  All of the YARN processes can be restarted with this Juju action.

• smoke-test

  Verify that YARN is working as expected by running a small (1MB) terasort.

• start-yarn

  All of the YARN processes can be started with this Juju action.

• stop-yarn

  All of the YARN processes can be stopped with this Juju action.

• teragen

  Generate data with teragen

  Params

  • indir string

    HDFS directory where generated data is stored

  • size integer

    The number of 100 byte rows, default to 1GB of data to generate

• terasort

  Runs teragen to generate sample data, and then runs terasort to sort that data

  Params

  • compression string

    Enable or Disable mapred output (intermediate) compression. LocalDefault will run with your current local hadoop configuration. Default means default hadoop deflate codec. One of: Gzip, BZip2, Snappy, Lzo, Default, Disable, LocalDefault These are all case sensitive.

  • indir string

    HDFS directory where generated data is stored

  • maps integer

    The default number of map tasks per job. 1-20

  • numtasks integer

    How many tasks to run per jvm. If set to -1, there is no limit.

  • outdir string

    HDFS directory where sorted data is stored

  • reduces integer

    The default number of reduce tasks per job. Typically set to 99% of the cluster's reduce capacity, so that if a node fails the reduces can still be executed in a single wave. Try 1-20

  • size integer

    The number of 100 byte rows, default to 1GB of data to generate and sort

• upgrade

  Upgrade this unit