0

I have a situation where I am using Spring Batch framework with JdbcCursorItemReader approach.

The problem is my query takes longer duration which is close to 1.2 to 1.5 mins to process records (work is in progress to optimize).

Chunk size is set to 60K.

In non-prod env, the task successfully processes the records to be pulled for a day (<60K) and completes.

Note:

  1. Total records in the table in non-prod: ~40-50K
  2. Total records in the table in prod: ~100-150K

But it appears that in prod, the task is not pulling entire set of records from the database (SQL SERVER)

For e.g. total no of records to be pulled for a day's data in DB is 25148. But the task pulled only 5148.

I am unable to replicate this issue in non-prod.

Any suggestions will be helpful.

Vikram Patil
  • 1
  • 2
  • Are there any errors in the debug log? What is different between your dev and prod environments? Without a [minimal example](https://stackoverflow.com/help/minimal-reproducible-example) that reproduces the issue, it would be difficult to tell you what could be wrong. – Mahmoud Ben Hassine Aug 11 '21 at 08:46
  • The key difference in dev and prod environments of database is the amount of data. Debug logs: The task completes its execution with no errors. Surprisingly only the last chunk of the data gets processed. – Vikram Patil Aug 11 '21 at 18:00

0 Answers0