Data getting missed while reading from azure event hub using spark streaming

[Rishi045]

Hi All,

I am facing an issue of data getting missed.

I am reading the data from azure event hub and after flattening the json data I am storing it in a parquet file and then using another databricks notebook to perform the merge operations on my delta table by adding some etl columns to it.

However in between somewhere the records are getting missed.

I have scheduled the job to run every hour.

Can someone please help me out with this.

[Rishi045]

As of now I am not having any foreachbatch in my code. 

I am performing dedup on entire data coming from event hub

[Hubert_Dudek1]

- In the EventHub, you can preview the event hub job using Azure Analitycs, so please first check are all records there

- Please set in Databricks that it is saved directly to the bronze delta table without performing any aggregation, just 1 to 1, and check if all records are there.

- Please consider using Delta Live Table for ingestion from Event Hub. It will make your live easier regarding monitoring stream, data quality, and performing full refresh when needed.