Hi everyone,

Following up on the Transaction Idempotency POC. Based on feedback from
Aleks on the ticket, we are pivoting away from patching individual modules
and moving towards a system-wide, generic interceptor approach aligned with
FINERACT-2169 <https://issues.apache.org/jira/browse/FINERACT-2169>.

I have just posted the revised technical architecture on the Jira ticket:
FINERACT-2485
<https://issues.apache.org/jira/projects/FINERACT/issues/FINERACT-2485?filter=allissues&orderby=level+ASC%2C+priority+DESC%2C+updated+DESC>

The proposed mechanics involve a Spring OncePerRequestFilter and a
dedicated DB table for atomic locking via unique constraints.

I am starting the implementation now but would appreciate any immediate red
flags regarding the database constraint approach for the atomic lock.

Best, Mohammed Saifulhuq

Reply via email to