Re: Writing a Zeppelin Interpreter in Scala

Hello Martin I faced the same issue as you when I tried to write the Cassandra interpreter using full Scala. You need to know that the magic of interpreter discovery is given by static { Interpreter.register(...) } There is no option to "emulate" a static block in Scala. Putting an init c

Writing a Zeppelin Interpreter in Scala

Hi, I'm trying to implement a Zeppelin interpreter in Scala. I have everything ready in the following form: package my.interpreter object MyInterpreter { Interpreter.register("MyInterpreter", classOf[MyInterpreter].getName) } class MyInterpreter(property: Properties) extends Interpreter(prope

Re: Could not find or load main class org.apache.spark.deploy.yarn.ExecutorLauncher

Try Zeppelin Tech Preview on HDP http://hortonworks.com/hadoop-tutorial/apache-zeppelin/ You should not get the error The YARN application launcher error is a Class not found exception when the classpath does not include right Zeppelin spark snapshot jar. On Thu, Nov 5, 2015 at 2:14 AM, Fengdong

Re: Zeppelin-0.5.5 and Spark-1.5.1

Yes, it should work On Wed, Nov 4, 2015 at 11:09 PM -0800, "Fengdong Yu" wrote: Hi Team Can I changed spark version to 1.5.1 and re-build? does that support spark-1.5.1 currently?

Re: Setting spark.app.name?

Hi moon, Yes, that’s exactly the change I was thinking of making. I’ll submit a PR in the next day or so. Thanks! Cheers, Craig > On Nov 5, 2015, at 9:30 AM, moon soo Lee wrote: > > It's easy to change default value of spark.app.name > based on environment variable

zeppelin startup warnings

Hello. I'm trying to run the master branch on CentOS6.5. My compile completes and the server start successfully, in the log file I see the following warnings. Question, are these expected? Also I'm uses spark with Mesos, I've followed the wiki instruction s but get the following stack trace o

Re: Setting spark.app.name?

It's easy to change default value of spark.app.name based on environment variable or jvm property. Changing https://github.com/apache/incubator-zeppelin/blob/master/spark/src/main/java/org/apache/zeppelin/spark/SparkInterpreter.java#L93 Would work. You can see how other setting reads the default v

Re: Setting spark.app.name?

So any interest in a pull request? Seems like a pretty straight-forward change, I’d be happy to do this including creating an env var and documenting it in zeppelin-env.sh.template. And seems an interesting feature when you’re running more than one zeppelin instance against a shared spark clus

Re: How to store Notebooks in S3

Thanks Josef and Victor. It is working. Thanks! 2015-11-04 12:36 GMT+01:00 Josef A. Habdank : > It worked for me. > > I have put a small help, which I might post to some blog at some point > (how to setup S3 access for Zeppelin on EMR): > > To have default EMR instance with Zeppelin store notebo

Re: Help in Zeppelin installation

Hi Pablo/Fengdong, I have used similar steps to resolve the build issues on multiple Ubuntu versions including 14.04 and 15.04. The only difference is that I used Maven 3.3.3, doesn't matter because the issue was due to Phantomjs and libfontconfig. And on Mac I was able to fix it by removing ~/.

Re: Help in Zeppelin installation

Hi guys, I don't know if it is the problem, but I succeed on my Ubuntu 14.04 VM after doing this: - Install libfontconfig - Uninstall maven, instead of using 'sudo apt-get install maven', I downloaded and I installed it manually, version 3.1.0 - Run 'mvn clean package -DskipTests' And everything

Re: Question about REST API to run a notebook

I would go the other way: The visualization and the back-end should not be strictly part of the Zeppelin core, while a generic interface to manage diverse pluggable backends though pluggable front-ends with a common core of managing notebooks and data structure would go a long way towards turning a

Could not find or load main class org.apache.spark.deploy.yarn.ExecutorLauncher

what’s possible ideas to debug this error. Spark-1.5.1, HDP2.2, Zeppelin-0.5.5 Zeppelin-0.5 does work.

Re: Help in Zeppelin installation

weird. I succeed on the Ubuntu VM, but always failed on Mac. > On Nov 5, 2015, at 5:07 PM, Corneau Damien wrote: > > It doesn't seem to be working. > However the good thing is that my new Ubuntu VM is having the same problem > than everybody. > So now I will be able to investigate that prob

Re: Help in Zeppelin installation

Ok, thanks Corneau. 2015-11-05 10:07 GMT+01:00 Corneau Damien : > It doesn't seem to be working. > However the good thing is that my new Ubuntu VM is having the same problem > than everybody. > So now I will be able to investigate that problem > > On Thu, Nov 5, 2015 at 9:47 AM, Pablo Torre > wr

Re: Zeppelin-0.5.5 and Spark-1.5.1

Thanks Moon > On Nov 5, 2015, at 4:54 PM, moon soo Lee wrote: > > If you built Zeppelin from master, yes it does support spark-1.5.1. > And you can even export SPARK_HOME to point 1.5.1 without rebuild Zeppelin. > > Thanks, > moon > > On Thu, Nov 5, 2015 at 4:09 PM Fengdong Yu

Re: "Do we have v0.5.5 binary package released"

It's not yet released. we're about to vote the 0.5.5 release. If you're interested, please test verify release candidate for 0.5.5 https://dist.apache.org/repos/dist/dev/incubator/zeppelin/ Best, moon On Thu, Nov 5, 2015 at 6:03 PM Jian Zhong wrote: > I tried to find 0.5.5 binary package releas

Re: Help in Zeppelin installation

It doesn't seem to be working. However the good thing is that my new Ubuntu VM is having the same problem than everybody. So now I will be able to investigate that problem On Thu, Nov 5, 2015 at 9:47 AM, Pablo Torre wrote: > Do you mean by using the PR that Courneau shared with us? > > 2015-11-0

Re: Spark read hive content

Hi, I think you'll need to manage get Zeppelin's Spark interpreter read hive-site.xml from it's classpath. Simplest way is placing hive-site.xml under conf/ directory of Zeppelin. Hope this helps. Best, moon On Sun, Nov 1, 2015 at 1:09 AM Ji, Hao Wei Jeffery wrote: > Hi, > > > > I’ve installe

"Do we have v0.5.5 binary package released"

I tried to find 0.5.5 binary package release. But can't find it. or only support build from source now? Thanks!

Re: Zeppelin-0.5.5 and Spark-1.5.1

If you built Zeppelin from master, yes it does support spark-1.5.1. And you can even export SPARK_HOME to point 1.5.1 without rebuild Zeppelin. Thanks, moon On Thu, Nov 5, 2015 at 4:09 PM Fengdong Yu wrote: > Hi Team > > Can I changed spark version to 1.5.1 and re-build? does that support > spa

Re: Including dependencies in Flink Interpreter

Hi, Here's related code in Zeppelin side. https://github.com/apache/incubator-zeppelin/blob/master/flink/src/main/java/org/apache/zeppelin/flink/FlinkInterpreter.java#L102 Argument placed in 3rd position is external jars. currently it's passing null. If user can set external jars in flink settin

Re: Help in Zeppelin installation

Do you mean by using the PR that Courneau shared with us? 2015-11-05 5:20 GMT+01:00 Fengdong Yu : > Did build on the MacBook? if so, please move to Ubuntu VM. I met the same > issue, but I finally succeed on the Ubuntu. > > > > On Nov 5, 2015, at 3:04 AM, Pablo Torre wrote: > > Hi Coernau, > >

Re: Zeppelin-web build failure

Hi Fengdong, What did you get it working? Thanks 2015-11-05 3:30 GMT+01:00 Fengdong Yu : > I finally succeed on the Ubuntu VM. not use your PR > > > > > On Nov 5, 2015, at 9:40 AM, Corneau Damien wrote: > > I guess I will have to make an Ubuntu VM then > > On Thu, Nov 5, 2015 at 2:38 AM, Fengd

Re: .spark.sql.AnalysisException: no such table ;

It can happen when you creates manually sqlContext instead of using Zeppelin provided one. Could you make sure you didn't create sqlContext manually? Thanks, moon On Fri, Oct 23, 2015 at 9:28 PM Tomas Hudik wrote: > Hi there, > i was following Zeppelin tutorial and code following snippet: > >

Re: Question about REST API to run a notebook

I am wondering if this does not make zeppelin an advance Spark-submit framework for multiple languages? As I see it, the visual representation of the result combined with the back-end language support is the core of zeppelin and REST api for submitting a notebook kind of kills it? How do we imagen