[jira] [Created] (ZEPPELIN-5887) Classloader issue when reading Parquet files.

["Daren Wong (Jira)" <ji...@apache.org>]

Daren Wong created ZEPPELIN-5887:
------------------------------------

             Summary: Classloader issue when reading Parquet files.
                 Key: ZEPPELIN-5887
                 URL: https://issues.apache.org/jira/browse/ZEPPELIN-5887
             Project: Zeppelin
          Issue Type: Bug
          Components: flink
    Affects Versions: 0.10.0
            Reporter: Daren Wong


I am trying to read a Parquet file into a table in Zeppelin but it fails with `java.lang.ClassNotFoundException: org.apache.hadoop.conf.Configuration`

Docker setup/Steps to reproduce:

```
docker run -u $(id -u) -p 8080:8080 -p 8081:8081 --rm -v /Users/xxx/Downloads/flink-1.13.6:/opt/flink -v /Users/xxx/Downloads/file.parquet:/opt/flink/data.parquet -e FLINK_HOME=/opt/flink --name zeppelin apache/zeppelin:0.10.0
```

flink-sql-parquet_2.12-1.13.6.jar is added to `/Users/xxx/Downloads/flink-1.13.6`.

I attempted to include the missing dependency by adding `flink-s3-fs-hadoop-1.13.6.jar` to `/Users/xxx/Downloads/flink-1.13.6` which then uncovers more missing dependencies `java.lang.ClassNotFoundException: org.apache.hadoop.mapreduce.lib.input.FileInputFormat`



--
This message was sent by Atlassian Jira
(v8.20.10#820010)