You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi,
We creating a hdfs sink connector based with a partition.field.name. This partition field is present in the middle of the avro data. HDFS Connector creates hive table structure as with the "PARTITIONED BY" as mentioned in the partition.field.name.
The issue here is, the location of the partition column in the parquet data file does not match with the hive table and this is causing wrong data to be read when accessing the particular column.
This seems to be working fine in the older versions of HDFS connect (5.5.2), was there any change made recently? We are seeing this issue on 10.1.1
The text was updated successfully, but these errors were encountered:
Hi,
We creating a hdfs sink connector based with a partition.field.name. This partition field is present in the middle of the avro data. HDFS Connector creates hive table structure as with the "PARTITIONED BY" as mentioned in the partition.field.name.
The issue here is, the location of the partition column in the parquet data file does not match with the hive table and this is causing wrong data to be read when accessing the particular column.
This seems to be working fine in the older versions of HDFS connect (5.5.2), was there any change made recently? We are seeing this issue on 10.1.1
The text was updated successfully, but these errors were encountered: