RE: Spark-Shell Issue

[Venkatesh]

Hi,

This is Venkatesh and I've been using your product for my hadoop and spark hands on practises on windows. I've to say, it is very helpful. Really cool integration of all hadoop tools into a unified platform.

But, lately have been experiencing issues with running scala commands in spark shell. Interestingly, I opened pyspark, and there it works perfectly.

Even a 1+1 command on a spark shell takes a while to execute, when I type enter, it enters character by character and takes so long to execute.

Tried manipulating executor cores and memory and nothing seem to work.

It'd be very helpful if you can take a look into the issue.

My email id is [e-mail]

Thanks,

Venkatesh Raman

[Boomigha Velayutham]

Hi Venkatesh, 

Thanks for using our Syncfusion products. 

Letters slowly getting typed scenario in Spark-Scala console occurs only in Windows 10 operating system. On checking the same, we would suggest to upgrade “jline” jar to latest one(2.14.6) in your Spark package. Can you please do the below steps to ensure that the issue gets resolved at your end? 

Steps: 

Step 1: Download the 2.14.6 version of jline jar from below location: 

              https://mvnrepository.com/artifact/jline/jline/2.14.6  

Step 2: Stop all the services (Refer below screenshot) 

                                  

Step 3: Replace the downloaded jar “jline-2.14.6” instead of “jline-2.12.1” in Spark Jars folder(..\Syncfusion\BigData\3.2.0.20\BigDataSDK\SDK\Spark\jars\jline-2.12.1). 

Step 4: Now start the services and make use of scala/pyspark. (Refer below screenshot) 

                                 

Please let us know, if you have any other concerns. 

Thanks, 

Boomigha 

[Pon Geetha A J]

Thanks for your timely help. The problem is solved. Really appreciate it 😊