[
https://issues.apache.org/jira/browse/YARN-8798?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16627542#comment-16627542
]
Antal Bálint Steinbach commented on YARN-8798:
----------------------------------------------
Hi [~tangzhankun] , [~wangda] ,
Submarine feature looks pretty interesting. This issue looks like a not soo
deep one, so I would be happy to contribute.
But first It would be great to be able to run at least some training jobs. In
the design doc the workflow part is in progress. Can you please give me some
guide or hints where to start?
If you are planning to do something like a getting started stuff I would be
happy to validate it.
> [Submarine] Job should not be submitted if "--input_path" option is missing
> ---------------------------------------------------------------------------
>
> Key: YARN-8798
> URL: https://issues.apache.org/jira/browse/YARN-8798
> Project: Hadoop YARN
> Issue Type: Sub-task
> Reporter: Zhankun Tang
> Assignee: Zhankun Tang
> Priority: Major
> Fix For: 3.2.0
>
>
> If a user doesn't set "–input_path" option, the job will still be submitted.
> Here is my command to run the job:
> {code:java}
> yarn jar
> $HADOOP_BASE_DIR/home/share/hadoop/yarn/hadoop-yarn-submarine-3.2.0-SNAPSHOT.jar
> job run \
> -verbose \
> -wait_job_finish \
> --env DOCKER_JAVA_HOME=/usr/lib/jvm/java-8-oracle \
> --env DOCKER_HADOOP_HDFS_HOME=/hadoop-3.2.0-SNAPSHOT \
> --name tf-job-001 \
> --docker_image tangzhankun/tensorflow \
> --worker_resources memory=4G,vcores=2 \
> --worker_launch_cmd "cd /cifar10_estimator && python cifar10_main.py
> --data-dir=%input_path% --job-dir=%checkpoint_path% --num-gpus=0
> --train-steps=5"{code}
> Due to lack of invalidity check, the job is still submitted. We should add a
> check on this.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
