[
https://issues.apache.org/jira/browse/SPARK-38677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17513536#comment-17513536
]
Thomas Graves commented on SPARK-38677:
---------------------------------------
Note, if you kill the python.daemon process while its hung, it will return to
pyspark console and have the right results.
I looked through commits in 3.2.1 and it appears that this was introduced by
https://issues.apache.org/jira/browse/SPARK-33277
Specifically commit
[https://github.com/apache/spark/commit/243c321db2f02f6b4d926114bd37a6e74c2be185]
At least I revert that commit and rebuilt and it then works. Also this did not
reproduce in standalone mode so it might just be a local mode issue.
[~ueshin] [~ankurdave] [~hyukjin.kwon]
> pyspark hangs in local mode running rdd map operation
> -----------------------------------------------------
>
> Key: SPARK-38677
> URL: https://issues.apache.org/jira/browse/SPARK-38677
> Project: Spark
> Issue Type: Bug
> Components: PySpark
> Affects Versions: 3.2.1, 3.3.0
> Reporter: Thomas Graves
> Priority: Blocker
>
> In spark 3.2.1 (spark 3.2.0 doesn't show this issue), pyspark will hang when
> running and RDD map operations and converting to a dataframe. Code is below
> to reproduce.
> Env:
> spark 3.2.1 local mode, just run {{./bin/pyspark --driver-memory XXXXG
> --driver-cores XXXX}}
> {{download dataset from here
> [https://rapidsai-data.s3.us-east-2.amazonaws.com/spark/mortgage.zip]}}
> {{just 200000 rows could reproduce the issue }}{{head -n 200000
> mortgage_eval_merged.csv > mortgage_eval_merged-small.csv}}{{{} but if the
> input dataset is small, such 50000 rows, it works well.{}}}{{{}{}}}run codes
> below:
> {code:java}
> path = "/XXXX/mortgage_eval_merged-small.csv" src_data =
> sc.textFile(path).map(lambda x:x.split(",")) column_list =
> ['c1','c2','c3','c4','c5','c6','c7','c8','c9','c10','c11','c12','c13','c14','c15','c16','c17','c18','c19','c20','c21','c22','c23','c24','c25','c26','c27','c28']
> df = spark.createDataFrame(src_data,column_list) print(df.show(1)){code}
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
