I want to read data from Teradata and write it into HDFS (HDP v2.4, scoop > 1.4.6) with parquet file format My first thought was to use tScoopImport component The problem is that the version of sqoop deployed in my cluster is 1.4.6 which is buged and it doesn't support custom sql. So I turned to parquet file : I created a spark job and I used tFileOutputParquet that works juste fine but he didn't support appending data to the same file even when I do set column partition ! So my question is : How can I append data in a parquet file written in HDFS ? Thanks you a lot dor your help.
