[
https://issues.apache.org/jira/browse/YARN-8569?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16602513#comment-16602513
]
Eric Yang commented on YARN-8569:
---------------------------------
{quote}As mentioned in the doc, the info appears with possible delay after
container launch. So containers need to add additional logics to get what is
the resources allocated to the container. However, such "sys info" for
individual containers should be created when container got launched. (Do you
think does it make sense which "/sys/fs/cgroups" created minutes after process
launched by OS?).{quote}
I plan to do another round to merge patch 1 and patch 5 to present resource
information prior to container launch. However, IP addresses will not be
available until docker reports them. Delay for obtaining IP is unavoidable.
{quote}And in implementation, AM should have ability to write files to an
arbitrary file instead of hard coded to "service.json". {quote}
This restriction is to prevent hiked YARN user from leaking more senstive
information like hdfs delegation token. We could also make a subdirectory in
nmPrviate to ensure that prefix path is not wide open for exploit.
> Create an interface to provide cluster information to application
> -----------------------------------------------------------------
>
> Key: YARN-8569
> URL: https://issues.apache.org/jira/browse/YARN-8569
> Project: Hadoop YARN
> Issue Type: Sub-task
> Reporter: Eric Yang
> Assignee: Eric Yang
> Priority: Major
> Labels: Docker
> Attachments: YARN-8569 YARN sysfs interface to provide cluster
> information to application.pdf, YARN-8569.001.patch, YARN-8569.002.patch,
> YARN-8569.003.patch, YARN-8569.004.patch, YARN-8569.005.patch
>
>
> Some program requires container hostnames to be known for application to run.
> For example, distributed tensorflow requires launch_command that looks like:
> {code}
> # On ps0.example.com:
> $ python trainer.py \
> --ps_hosts=ps0.example.com:2222,ps1.example.com:2222 \
> --worker_hosts=worker0.example.com:2222,worker1.example.com:2222 \
> --job_name=ps --task_index=0
> # On ps1.example.com:
> $ python trainer.py \
> --ps_hosts=ps0.example.com:2222,ps1.example.com:2222 \
> --worker_hosts=worker0.example.com:2222,worker1.example.com:2222 \
> --job_name=ps --task_index=1
> # On worker0.example.com:
> $ python trainer.py \
> --ps_hosts=ps0.example.com:2222,ps1.example.com:2222 \
> --worker_hosts=worker0.example.com:2222,worker1.example.com:2222 \
> --job_name=worker --task_index=0
> # On worker1.example.com:
> $ python trainer.py \
> --ps_hosts=ps0.example.com:2222,ps1.example.com:2222 \
> --worker_hosts=worker0.example.com:2222,worker1.example.com:2222 \
> --job_name=worker --task_index=1
> {code}
> This is a bit cumbersome to orchestrate via Distributed Shell, or YARN
> services launch_command. In addition, the dynamic parameters do not work
> with YARN flex command. This is the classic pain point for application
> developer attempt to automate system environment settings as parameter to end
> user application.
> It would be great if YARN Docker integration can provide a simple option to
> expose hostnames of the yarn service via a mounted file. The file content
> gets updated when flex command is performed. This allows application
> developer to consume system environment settings via a standard interface.
> It is like /proc/devices for Linux, but for Hadoop. This may involve
> updating a file in distributed cache, and allow mounting of the file via
> container-executor.
--
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
