Duplicate Data with BlueGreen Deployment

Our usecase is streaming data from kafka --> processing them in flink --> and saving the data in a database.
With the blue green deployment, i assume while both blue and green are up both would be sending data in my db. Is that correct?
How would you recommend preventing this to happen so my data is still consistent.

No files or tests are mentioned. Start by reviewing the operator's blue-green deployment behavior for Kafka-to-Flink-to-database workloads, then document whether both deployments can write concurrently and what consistency-preserving approach is recommended. Done means the deployment behavior and prevention guidance are clearly recorded.

Written by the indexing model from the issue text.