Situations

• There is a service, which is providing data with many columns but we require only 10 columns. The service fetches data for many joins. We have the database available with us. We can directly use hibernate calls for getting data which can help in reducing the time complexity.
• Data Inconsistency is to be handled when one services and other updates, the record
• We can save the progress data in non-relational database like Mango DB
• Request, service will broadcast and error to topic in Kafka Which will be consumed by other services to perform specific cleanup.
• We can also have schedule running that will check for transactions failed, and will retry them and if A transaction is not success, then we retry For specific attempts and then leave it And send response in a log.