Are you sure it's deadlock? print the thread dump (from kill -QUIT) of the thread(s) that are deadlocked, I suppose, to show where the issue is. It seems unlikely that a logging thread would be holding locks that the app uses.
On Fri, Nov 28, 2014 at 4:01 PM, Charles <charles...@cenx.com> wrote: > We create spark context in an application running inside wildfly container. > When spark context is created, we see following entires in the wildfly log. > After the log4j-default.properties is loaded, every entry from spark is > printed out twice. And after running for a while, we start to see deadlock > between spark logging thread and wildfly logging thread. > > Can I control the spark logging in the driver application? How can I turn > it off in the driver application? How can I control the level of spark logs > in the driver application? > > 2014-11-27 14:39:26,719 INFO [akka.event.slf4j.Slf4jLogger] > (spark-akka.actor.default-dispatcher-4) Slf4jLogger started > 2014-11-27 14:39:26,917 INFO [Remoting] > (spark-akka.actor.default-dispatcher-2) Starting remoting > 2014-11-27 14:39:27,719 INFO [Remoting] > (spark-akka.actor.default-dispatcher-2) Remoting started; listening on > addresses :[akka.tcp://spark@172.32.1.12:43918] > 2014-11-27 14:39:27,733 INFO [Remoting] > (spark-akka.actor.default-dispatcher-2) Remoting now listens on addresses: > [akka.tcp://spark@172.32.1.12:43918] > 2014-11-27 14:39:27,892 INFO [org.apache.spark.SparkEnv] (MSC service > thread 1-16) Using Spark's default log4j profile: > org/apache/spark/log4j-defaults.properties > 2014-11-27 14:39:27,895 ERROR [stderr] (MSC service thread 1-16) 14/11/27 > 14:39:27 INFO SparkEnv: Using Spark's default log4j profile: > org/apache/spark/log4j-defaults.properties > 2014-11-27 14:39:27,896 INFO [org.apache.spark.SparkEnv] (MSC service > thread 1-16) Registering BlockManagerMaster > 2014-11-27 14:39:27,896 ERROR [stderr] (MSC service thread 1-16) 14/11/27 > 14:39:27 INFO SparkEnv: Registering BlockManagerMaster > 2014-11-27 14:39:28,041 INFO [org.apache.spark.storage.DiskBlockManager] > (MSC service thread 1-16) Created local directory at > /tmp/spark-local-20141127143928-d33c > 2014-11-27 14:39:28,041 ERROR [stderr] (MSC service thread 1-16) 14/11/27 > 14:39:28 INFO DiskBlockManager: Created local directory at > /tmp/spark-local-20141127143928-d33c > 2014-11-27 14:39:28,055 INFO [org.apache.spark.storage.MemoryStore] (MSC > service thread 1-16) MemoryStore started with capacity 4.3 GB. > 2014-11-27 14:39:28,055 ERROR [stderr] (MSC service thread 1-16) 14/11/27 > 14:39:28 INFO MemoryStore: MemoryStore started with capacity 4.3 GB. > 2014-11-27 14:39:28,117 INFO [org.apache.spark.network.ConnectionManager] > (MSC service thread 1-16) Bound socket to port 34018 with id = > ConnectionManagerId(ip-172-32-1-12,34018) > 2014-11-27 14:39:28,118 ERROR [stderr] (MSC service thread 1-16) 14/11/27 > 14:39:28 INFO ConnectionManager: Bound socket to port 34018 with id = > ConnectionManagerId(ip-172-32-1-12,34018) > 2014-11-27 14:39:28,162 INFO [org.apache.spark.storage.BlockManagerMaster] > (MSC service thread 1-16) Trying to register BlockManager > 2014-11-27 14:39:28,163 ERROR [stderr] (MSC service thread 1-16) 14/11/27 > 14:39:28 INFO BlockManagerMaster: Trying to register BlockManager > 2014-11-27 14:39:28,181 INFO > [org.apache.spark.storage.BlockManagerMasterActor$BlockManagerInfo] > (spark-akka.actor.default-dispatcher-3) Registering block manager > ip-172-32-1-12:34018 with 4.3 GB RAM > 2014-11-27 14:39:28,185 ERROR [stderr] > (spark-akka.actor.default-dispatcher-3) 14/11/27 14:39:28 INFO > BlockManagerMasterActor$BlockManagerInfo: Registering block manager > ip-172-32-1-12:34018 with 4.3 GB RAM > > > > -- > View this message in context: > http://apache-spark-user-list.1001560.n3.nabble.com/Deadlock-between-spark-logging-and-wildfly-logging-tp20009.html > Sent from the Apache Spark User List mailing list archive at Nabble.com. > > --------------------------------------------------------------------- > To unsubscribe, e-mail: user-unsubscr...@spark.apache.org > For additional commands, e-mail: user-h...@spark.apache.org > --------------------------------------------------------------------- To unsubscribe, e-mail: user-unsubscr...@spark.apache.org For additional commands, e-mail: user-h...@spark.apache.org
