No Help from Nvidia as usual. They are total trash.
UFF and Caffe Parser have been deprecated from TensorRT 7 onwards, hence request you to try ONNX parser.
Please check the below link for the same.
Nvidia is trash.
Sorry for the inconvenience.
Please share with us the issue repro executable script and ONNX model to try from our end for better debugging.
Meanwhile please refer official sample on INT8 calibration and make sure your script has been written correctly.