Liyin Tang created SPARK-13580:
----------------------------------
Summary: Driver makes no progress after failed to remove broadcast
on Executor
Key: SPARK-13580
URL: https://issues.apache.org/jira/browse/SPARK-13580
Project: Spark
Issue Type: Bug
Components: Streaming
Affects Versions: 1.5.2
Reporter: Liyin Tang
>From Driver's log: it failed to remove broadcast data due to RPC timeout
>exception from executor #11. And it also failed to get thread dump from
>executor #11 due to akka.actor.ActorNotFound exception.
After that, driver waited for executor #11 to finish one task for that job. All
the other tasks are finished for that job.
However, from the executor#11's log, it didn't get that task (it got 9 other
tasks and finished them)
Since then, there is no progress in the streaming job.
I have attached the driver's log and jstack, executor's jstack.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]