[
https://issues.apache.org/jira/browse/SPARK-23840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16422242#comment-16422242
]
Uri Goren commented on SPARK-23840:
-----------------------------------
This error does not occur locally,
I have checked why, and apparently I was running pyspark 2.3 remotely, and
pyspark 2.2.1 locally.
After downgrading the pyspark version, everything works as expected.
The error message occurs in `worker.py` in a `loads` function.
If you need, I can regenerate the exception with pyspark 2.3.0 on EMR 5.11, by
using basic rdd operations on parquet data.
I can check with my employer if I can share a cluster / data set with you if
it's absolutely necessary.
> PySpark error when converting a DataFrame to rdd
> ------------------------------------------------
>
> Key: SPARK-23840
> URL: https://issues.apache.org/jira/browse/SPARK-23840
> Project: Spark
> Issue Type: Bug
> Components: PySpark
> Affects Versions: 2.3.0
> Reporter: Uri Goren
> Priority: Major
>
> I am running code in the `pyspark` shell on an `emr` cluster, and
> encountering an error I have never seen before...
> This line works:
> spark.read.parquet(s3_input).take(99)
> While this line causes an exception:
> spark.read.parquet(s3_input).rdd.take(99)
> With
> > TypeError: 'int' object is not iterable
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
