Hi, I think at this point it might be worth creating an issue to track this and bring it to attention dev@
Thanks! On Wed, Sep 2, 2015 at 12:14 PM, zhao selery <seleryz...@gmail.com> wrote: > I run scala code from zeppelin notebook, when I look at spark log from > spark/work/app-20150901193944-0015/0/stderr > Log file looks like: > SLF4J: Class path contains multiple SLF4J bindings. > SLF4J: Found binding in > [jar:file:/home/zhaoxq/Downloads/zeppelin2/incubator-zeppelin/interpreter/spark/zeppelin-spark-0.6.0-incubating-SNAPSHOT.jar!/org/slf4j/impl/StaticLoggerBinder.class] > SLF4J: Found binding in > [jar:file:/home/zhaoxq/Downloads/zeppelin2/incubator-zeppelin/zeppelin-interpreter/target/lib/slf4j-log4j12-1.7.10.jar!/org/slf4j/impl/StaticLoggerBinder.class] > SLF4J: Found binding in > [jar:file:/home/zhaoxq/spark/assembly/target/scala-2.10/spark-assembly-1.5.0-SNAPSHOT-hadoop2.2.0.jar!/org/slf4j/impl/StaticLoggerBinder.class] > SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an > explanation. > SLF4J: Actual binding is of type [org.slf4j.impl.Log4jLoggerFactory] > > > But if I run same code from spark-shell, the log file containts correct > spark log info: > 15/09/01 19:31:58 INFO CoarseGrainedExecutorBackend: Registered signal > handlers for [TERM, HUP, INT] > 15/09/01 19:31:58 WARN NativeCodeLoader: Unable to load native-hadoop > library for your platform... using builtin-java classes where applicable > 15/09/01 19:31:59 INFO SecurityManager: Changing view acls to: zhaoxq > 15/09/01 19:31:59 INFO SecurityManager: Changing modify acls to: zhaoxq > 15/09/01 19:31:59 INFO SecurityManager: SecurityManager: authentication > disabled; ui acls disabled; users with view permissions: Set(zhaoxq); users > with modify permissions: Set(zhaoxq) > 15/09/01 19:31:59 INFO Slf4jLogger: Slf4jLogger started > ................. > ................ > > > Also, if I rebuild zeppelin with remote false mode, which is: > change setting.getOption().setRemote(false); in > org.apache.zeppelin.interpreter.InterpreterFactory, > spark worker log is same as running from spark-shell, which is what I want. > > > So, my question is how can I get useful spark worker log with zeppelin > remote mode? Thanks. > -- -- Kind regards, Alexander.
