[
https://issues.apache.org/jira/browse/YARN-4205?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15525225#comment-15525225
]
Gour Saha commented on YARN-4205:
---------------------------------
YARN-4692 also needs total execution timeout of application. It does not need
queue or state-store timeouts. The only additional thing it needs, is the
monitor start time to provide options like LAUNCH_FIRST and LAUNCH_EVERYTIME on
top of SUBMISSION.
YARN-4692 is for long running or semi long-running services. Long running
services are meant to run forever. Semi long-running services are meant to run
for several hours or few days or weeks. For the semi long-running usecases, we
have applications like a CI (continuous-integration) app or a System Test app,
which typically needs several hours to finish its job. CI app or System Test
app owners are not going to be happy that their app did not run at all, even
though it was given a timeout of 10 hours because YARN got a chance to allocate
resource to it at the 9:59 hour mark since submission. That is why we need
LAUNCH_FIRST.
LAUNCH_EVERYTIME will cover scenarios like say a load test app needs to run
end-to-end for at least 2 days straight to certify a product. If such an app is
pre-empted at say the 47 hour mark, it needs a fresh 48 hour lifetime the next
time it is re-launched.
> Add a service for monitoring application life time out
> ------------------------------------------------------
>
> Key: YARN-4205
> URL: https://issues.apache.org/jira/browse/YARN-4205
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: scheduler
> Reporter: nijel
> Assignee: Rohith Sharma K S
> Attachments: 0001-YARN-4205.patch, 0002-YARN-4205.patch,
> 0003-YARN-4205.patch, 0004-YARN-4205.patch, 0005-YARN-4205.patch,
> 0006-YARN-4205.patch, YARN-4205_01.patch, YARN-4205_02.patch,
> YARN-4205_03.patch
>
>
> This JIRA intend to provide a lifetime monitor service.
> The service will monitor the applications where the life time is configured.
> If the application is running beyond the lifetime, it will be killed.
> The lifetime will be considered from the submit time.
> The thread monitoring interval is configurable.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
