spark-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Michael Armbrust <mich...@databricks.com>
Subject Re: [SQL] HiveThriftServer2 failure detection
Date Wed, 19 Nov 2014 18:48:14 GMT
This is not by design.  Can you please file a JIRA?

On Wed, Nov 19, 2014 at 9:19 AM, Yana Kadiyska <yana.kadiyska@gmail.com>
wrote:

> Hi all, I am running HiveThriftServer2 and noticed that the process stays
> up even though there is no driver connected to the Spark master.
>
> I started the server via sbin/start-thriftserver and my namenodes are
> currently not operational. I can see from the log that there was an error
> on startup:
>
> 14/11/19 16:32:58 ERROR HiveThriftServer2: Error starting HiveThriftServer2
>
> and that the driver shut down as expected:
>
> 14/11/19 16:32:59 INFO SparkUI: Stopped Spark web UI at http://myip:4040
> 14/11/19 16:32:59 INFO DAGScheduler: Stopping DAGScheduler
> 14/11/19 16:32:59 INFO SparkDeploySchedulerBackend: Shutting down all executors
> 14/11/19 16:32:59 INFO SparkDeploySchedulerBackend: Asking each executor to shut down
> 14/11/19 16:33:00 INFO MapOutputTrackerMasterActor: MapOutputTrackerActor stopped!
> 14/11/19 16:33:00 INFO MemoryStore: MemoryStore cleared
> 14/11/19 16:33:00 INFO BlockManager: BlockManager stopped
> 14/11/19 16:33:00 INFO BlockManagerMaster: BlockManagerMaster stopped
> 14/11/19 16:33:00 INFO SparkContext: Successfully stopped SparkContext
>
> ​
> However, when I try to run start-thriftserver.sh again I see an error
> message that the process is already running and indeed there is a process
> with that PID:
>
> root     32334     1  0 16:32 ?        00:00:00 /usr/local/bin/java org.apache.spark.deploy.SparkSubmitDriverBootstrapper
--class org.apache.spark.sql.hive.thriftserver.HiveThriftServer2 --master spark://myip:7077
--conf -spark.executor.extraJavaOptions=-verbose:gc -XX:-PrintGCDetails -XX:+PrintGCTimeStamps
spark-internal --hiveconf hive.root.logger=INFO,console
>
> ​
>  Is this a bug or design decision -- I am upgrading from Shark and we had
> scripts that monitor the driver and restart on failure. Here it seems that
> we would not be able to restart even though the driver died?
>

Mime
View raw message