[
https://issues.apache.org/jira/browse/SPARK-21502?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Marcelo Vanzin resolved SPARK-21502.
------------------------------------
Resolution: Fixed
Assignee: Stavros Kontopoulos
Fix Version/s: 2.3.0
> --supervise causing frameworkId conflicts in mesos cluster mode
> ---------------------------------------------------------------
>
> Key: SPARK-21502
> URL: https://issues.apache.org/jira/browse/SPARK-21502
> Project: Spark
> Issue Type: Bug
> Components: Mesos
> Affects Versions: 2.2.0
> Reporter: Stavros Kontopoulos
> Assignee: Stavros Kontopoulos
> Fix For: 2.3.0
>
>
> This issue is described here: https://dcosjira.atlassian.net/browse/SPARK-8
> The --supervise option is actually broken in mesos cluster mode.
> The error message you get when you re-launch a driver is:
> 17/05/16 13:26:09 ERROR MesosCoarseGrainedSchedulerBackend: Mesos error:
> Framework has been removed
> Exception in thread "Thread-13" org.apache.spark.SparkException: Exiting due
> to error from cluster scheduler: Framework has been removed
> at
> org.apache.spark.scheduler.TaskSchedulerImpl.error(TaskSchedulerImpl.scala:459)
> at
> org.apache.spark.scheduler.cluster.mesos.MesosCoarseGrainedSchedulerBackend.error(MesosCoarseGrainedSchedulerBackend.scala:569)
> I0516 13:26:09.003342 92 sched.cpp:2021] Asked to abort the driver
> I0516 13:26:09.003362 92 sched.cpp:1217] Aborting framework
> '1b9cb705-2e80-4354-8f3b-a4cffac7dac3-0003-driver-20170516131736-0004'
> 17/05/16 13:26:09 INFO MesosCoarseGrainedSchedulerBackend: driver.run()
> returned with code DRIVER_ABORTED
> 17/05/16 13:26:09 ERROR SparkContext: Error initializing SparkContext.
> org.apache.spark.SparkException: Error starting driver, DRIVER_ABORTED
> at
> org.apache.spark.scheduler.cluster.mesos.MesosSchedulerUtils$$anon$1.run(MesosSchedulerUtils.scala:125)
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
