Danny Chan comments

Results 402 comments of


                                            Danny Chan

[SUPPORT] Archive operation always release lock on the timeline when try lock failed

Okay, it would be great if you can fire a fix for it.

[HUDI-7576] Improve efficiency of getRelativePartitionPath, reduce computation of partitionPath in AbstractTableFileSystemView

@the-other-tim-brown Can you fix the Azure CI failure?

[HUDI-1517][HUDI-6758][HUDI-6761] Adding support for per-logfile marker to track all log files added by a commit and to assist with rollbacks

> I would caution against landing it in 0.14.0. -1 for landing it in 0.14.0.

[HUDI-1517][HUDI-6758][HUDI-6761] Adding support for per-logfile marker to track all log files added by a commit and to assist with rollbacks

Take a glimpse of the description and I'm not very approval about these changes: 1. Add extraneous log files to the commit metadata either for writer commit or repetitive rollback...

[SUPPORT] throw "java.lang.NoSuchMethodError: org.apache.hadoop.hdfs.client.HdfsDataInputStream.getReadStatistics()

Looks like a Hadoop jar conflict.

[SUPPORT] Flink-Hudi Unable to use Hudi metadata with S3

yeah, the mdt is not recommended for Flink 0.13.x now, we are striving to make it production ready for 1.x release.

[SUPPORT] Flink-Hudi Unable to use Hudi metadata with S3

1.0-beta2 already released out, you can have a try, just be caution that 1.x has some format incompatibility with 0.x, we are trying to resolve those in 1.0 GA release.

[SUPPORT] Flink-Hudi Unable to use Hudi metadata with S3

You can try other lock provider impl.

[SUPPORT] java.lang.ClassCastException: class org.apache.spark.sql.catalyst.expressions.UnsafeRow cannot be cast to class org.apache.spark.sql.vectorized.ColumnarBatch

It looks like a known issue reported in: https://github.com/apache/hudi/issues/9305

[SUPPORT] java.lang.ClassCastException: class org.apache.spark.sql.catalyst.expressions.UnsafeRow cannot be cast to class org.apache.spark.sql.vectorized.ColumnarBatch

> Do you think the issue is related to how the metadata table is saved? In other words, is metadata not supported with Spark 3.3.2? It is supported, can you...