Hello @lqthinguyen
When a Qlik Replicate task takes a long time to stop, it is usually because it is waiting for a response from the database.
Example: If a Change Data Capture (CDC) query takes 20 minutes to get a response from the database, the Replicate task has to wait for those 20 minutes before it can stop completely.
The duration of database queries directly affects the stopping time of the Replicate task.
Qlik Replicate creates multiple connections to the database. Some of these connections may remain idle for a long time.
Databases or firewalls might drop these idle connections, which can cause Replicate to take a long time to stop as it tries to close these connections gracefully.
Additionally, If Replicate is waiting for any process in the target/source (like a lock, file transfer, or network reply), the task will take longer to stop.
When you stop a task, it stops connections from the source. The stopping process can take up to 30 minutes, depending on the state and amount of data being processed at the time.
Please note , there is an internal 30-minute clock that governs the maximum time a task can take to stop and close connections.
This does not mean Replicate will always wait 30 minutes; if there is little activity, the task will stop faster.
May be as a workaround , If you do not want to wait for the entire stopping process, you can take the following steps:
Check Logs:
Look for the log entry "[SORTER ]I: Final saved task state. Stream position "
This indicates that the task has reached a state where it is safe to terminate it.
Kill the Task via task manager if its windows environment or kill the PID if its Linux
Regards
Arun
