You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@spark.apache.org by Sergio Ramírez <sr...@ugr.es> on 2016/04/14 12:18:50 UTC

[Streaming] Infinite delay when stopping the context

Hello:

I use the stop method in my streaming programs to finish the executions 
of my experiments. However, despite of getting these messages:
/
//16/04/14 12:03:39 INFO JobGenerator: Stopping JobGenerator immediately//
//16/04/14 12:03:39 INFO RecurringTimer: Stopped timer for JobGenerator 
after time 1460628219000//
//16/04/14 12:03:39 INFO JobGenerator: Stopped JobGenerator//
//16/04/14 12:03:39 INFO ReceivedBlockTracker: Deleting batches 
ArrayBuffer()

/The thread associated to this app never ends completely. If I look for 
this process in the OS, I can see that it is still running. I get the 
same problem with both version of the stop function (gracefully and 
abrupt). The spark version used is 1.6.1.

Any clue about this problem?

Regards,

Sergio R.