Re: No active SparkContext black hole

Hi Jeff, Thanks for your response. This happens during long running yarn-client Spark jobs, everything is going fine, lots of output in the interpreter log, then we see a failed sent message. INFO [2016-10-05 17:31:49,586] ({spark-dynamic-executor-allocation} Logging.scala[logInfo]:58) - Request

Problem with local jars in Zeppelin 0.6.1 with Spark 2.0.0

Hi everyone, I'm using Zeppelin 0.6.1 with Spark 2.0.0 on Yarn (with master configured as yarn-client). I have couple of jars stored locally which I'd like to use in Spark interpreter. I've put these jars in `spark-defaults.conf` under `spark.jars` (as described in https://zeppelin.apache.org/do c

Re: Problem with local jars in Zeppelin 0.6.1 with Spark 2.0.0

What are you doing in Spark that causes Zeppelin to throw that error. Have you loaded Jars to the interpreter as dependencies as well? [image: Inline images 1] HTH Dr Mich Talebzadeh LinkedIn * https://www.linkedin.com/profile/view?id=AAEWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw

Re: Getting java.lang.NoSuchFieldException: MODULE$ in number of scenarios

anyone having similar issue? On Thu, Oct 6, 2016 at 12:00 PM, Nirav Patel wrote: > I forgot to mention other scenario where I get same error. It's during use > of `org.apache.spark.mllib.evaluation.MulticlassMetrics` > > It fails on following call > > val predictionMatrix = new MulticlassMetrics

Re: Problem with local jars in Zeppelin 0.6.1 with Spark 2.0.0

Thanks a lot Mich! Somehow I missed that it's possible to put filesystem path in interpreter config. Regards, Arkadiusz 2016-10-10 19:27 GMT+02:00 Mich Talebzadeh : > What are you doing in Spark that causes Zeppelin to throw that error. > > Have you loaded Jars to the interpreter as dependencies