Hi, I am using HDP 2.6.5, Spark 2.3 and Hive 1.2. I am running a pyspark program and in this program a method is being called multiple times with different date ranges (begin and end date). This is being called sequentially in a loop. However, I have modified it to call this method in parallel using x number of threads with each thread having the different start and end dates. I am observing some weird behavior in the sense that, the threads do seem to append data into this table but however, for every run of the program, the counts are differing and not the same. I am using a hive managed table (not external table) which is in orc format. Data is not partitioned.