Do you mind file an issue? I think i can quickly make a patch. Thanks, moon
On Sun, May 29, 2016 at 5:53 PM Polina Marasanova < polina.marasan...@quantium.com.au> wrote: > Thanks for the response > > I added 'args' property at the interpreter tab like this: > > spark %spark (default) , %pyspark , %sql , %dep edit restart remove > Properties > > name value > args -Xmax-classfile-name=128 > > But unfortunately it's still reproducible. Any other ways how to do it? > ________________________________________ > From: moon soo Lee [m...@apache.org] > Sent: Saturday, 28 May 2016 6:00 AM > To: users@zeppelin.apache.org > Subject: Re: Getting 'File name too long' error when running scala code in > Zeppelin > > Hi, > > You can see 'args' property on spark interpreter setting, in interpreter > menu. Could you try there and see if it works? > > Thanks, > moon > > On Fri, May 27, 2016 at 12:18 AM Polina Marasanova < > polina.marasan...@quantium.com.au<mailto:polina.marasan...@quantium.com.au>> > wrote: > Hello, > > Sometimes running scala code in Zeppelin I'm getting an error 'File name > too long'. To keep working in Notebook I need to restart the interpreter > after it. Seems that it's a Scala bug which can be fixed by this workaround > https://issues.apache.org/jira/browse/SPARK-4820 > Could you please advice me what is the right way to apply this option? > I already tried this to add in spark-defaults.conf this: > spark.driver.extraJavaOptions -Xmax-classfile-name=128 > And it doesn't help. > > Thank you > > Cheers > Polina Marasanova > >
