Nifi 1.9.2, single node, single cluster.
I am using 5 ExecuteSql processor to bulk collect records from Oracle DB.
(each select query suppose to pick separate set of one fifth of total volume, logic is based on MOD function).
Predecessor for all the executesql processors is same (generateflowfile processor).
ExecuteSql processors are configured as ("TIMER Driven", concurrent task=40, run schedule=30 sec, "maximum rows per flowfile"=500 and "output batch size"=1).
We are starting the nifi flow, then generateflowfile processor is generating one flow file which is triggering executeSql processors.
But randomly in some of the runs, data is getting duplicated.
Not sure if this is the issue with concurrent tasks or timer driven property.
What was the solution for this issue in not having duplicate records when using ExecuteSql in Nifi. how to avoid the duplicate records. My flow is ExecuteSql --> PutDatabaseREcords.