Hi, I am facing issue accessing one of the child json attribute while forming SQL in QureyRecord processor. PFB details   here is my Avro schema added for JSON reader and writter. { "name": "MyClass", "type": "record", "namespace": "com.acme.avro", "fields": [ { "name": "labels", "type": { "name": "labels", "type": "record", "fields": [ { "name": "__name__", "type": "string" }, { "name": "cucsMemoryUnitInstanceId", "type": "string" }, { "name": "instance", "type": "string" }, { "name": "job", "type": "string" }, { "name": "monitor", "type": "string" }, { "name": "site_identifier", "type": "string" } ] } }, { "name": "name", "type": "string" }, { "name": "timestamp", "type": "string" }, { "name": "value", "type": "string" }, { "name": "producedAt", "type": "string" } ] }   now the SQL quary for first level attribute works fine.    like SELECT * from FLOWFILE where name = 'xyz'   or  SELECT * from FLOWFILE where producedAt = 'dd-mm-yyyy' however If I have to access attribute of nested JSON it wont work   e.g SELECT * from FLOWFILE where site_identifier = 'xyz'    --> This throws the error. ... View more

Hi, We are trying to fetch the data from one of our kafka cluster (200000 messages per seconds) where we have 19 partition on the topic and using nifi (cluster of 4 hosts) inbuilt processor (ConsumeKafka_2_6) however we are observing delay in processing messages in the processor (linage time of 7 sec per message). We tried using increase concurrent task to match up with the partition as well but it did not work. however if we use demarcation strategy (to club multiple messages into single flowfile) it is working fine but then we need individual messages for data transformation.   The other processor which does the JSON evaluation also have similar problem as it can not handle large incoming data from the cosumekafka processor.   Notes : Backpressure threshold is 100000 file and size threshold is 1 GB PFB reference SS of the processor configured.   Evaluate Json path config:  (Run duration 0.05 ms,  concurrent task-4 per node )       ... View more