Lambda queries do not reflect the database state

Question

Hi,  
  
I have a lambda that connects to an Amazon Aurora RDS instance and gets data from it. In the database, I have two tables, Student and Processed. My lambda query returns 5 students from the student table that have not been processed,  i.e., their ids do not appear in the Processed table. I'm noticing that when I make an update to the Processed table the changes are not immediately reflected in the lambda result. That is, the lambda still returns students that have been marked as processed. Even though the data in the database has changed. I have verified the updated state of the database manually. Is it possible that the lambda is caching somewhere? I'm executing the lambda from the AWS console so there are no layers there.  
  
I also noticed that if I make a minor change to the lambda code, save it, and then execute it, I get the correct results. Thanks for any help.

Answer

I figured it out. I was creating the session object outside the lambda handler which caused it to be cached. I moved session creation inside the handler and everything works now.

Lambda queries do not reflect the database state

Relevant questions

Serverless lambda rest api not updating database properly

I have lots of tables that are stuck in the "Pending Records" Validation state and I can't get them to progress

Why the LoginFailures metric showing no data for RDS MySQL instance?

Internal Server Error from API Gateway when sending queries through gateway to Lambda function connected to RDS database

Why can't I find my RDS database from my lambda function?

Handler error when connecting lambda function to RDS database

Database in Amazon Glue

Error establishing a database connection with RDS and OpenLiteSpeed AIMs