[
https://issues.apache.org/jira/browse/SPARK-6415?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15044110#comment-15044110
]
Yoel Amram commented on SPARK-6415:
-----------------------------------
It would be useful to let the application decide when to fail-fast, I can see
several relevant use cases:
1. one or more workers are out of memory (could be unbalanced load causing only
one/several partitions to fail).
2. all workers fail because of unreachable destination (e.g. network issues
with target database).
Also, maybe having a similar configuration to spark.task.maxFailures (i.e.
spark.job.maxFailures) to retry job execution several times (with possible
backoff period) before exiting the application.
> Spark Streaming fail-fast: Stop scheduling jobs when a batch fails, and kills
> the app
> -------------------------------------------------------------------------------------
>
> Key: SPARK-6415
> URL: https://issues.apache.org/jira/browse/SPARK-6415
> Project: Spark
> Issue Type: Improvement
> Components: Streaming
> Reporter: Hari Shreedharan
>
> Of course, this would have to be done as a configurable param, but such a
> fail-fast is useful else it is painful to figure out what is happening when
> there are cascading failures. In some cases, the SparkContext shuts down and
> streaming keeps scheduling jobs
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
