vinothchandar commented on issue #877: java.lang.NoSuchMethodError: io.netty.buffer.PooledByteBufAllocator.metric()Lio/netty/buffer/PooledByteBufAllocatorMetric; URL: https://github.com/apache/incubator-hudi/issues/877#issuecomment-528454357 @wannaberich the integration is documented here http://hudi.apache.org/querying_data.html I dont have experience with HDP myself. But for Zeppelin itself, I have made it work by including the spark bundle on the `spark.jars` property of the spark interpreter. Orthogonally, what I was saying is, you can try building teh bundles off this PR #873 and it should be way painless.. Some of the jar issues could go away . However this seems like ``` Caused by: java.lang.NoSuchMethodError: org.apache.hadoop.fs.FSDataOutputStream: method <init>(Ljava/io/OutputStream;)V not found ``` HDFS 3 being not backwards compatible? They have removed the constructor here https://github.com/apache/hadoop/blob/a55d6bba71c81c1c4e9d8cd11f55c78f10a548b0/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/FSDataOutputStream.java#L77 If there a way you can stick to a lower HADOOP & HIVE versions? Most cloud providers (EMR, Data proc) are all on Hive 2.x and Hadoop 2.x still. Hudi has not yet been tests for Hadoop 3 or Hive 3 (we know both of them have bad breaking changes)
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
