Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hi,
I have a theoretical scenario. Task is from SQL to SQL using MS-CDC.
1. To simulate the corruption of the source MS-CDC tables (the tables that store changes) in a SQL Server database, our client will create a QR task that runs in Full Load and CDC mode, with multiple tables. The task is started and it is left in the running state.
2. While the task is running, one of the MS-CDC tables in the source database is deleted, and the QR task fails and stops.
My question is, if we manually restore the deleted CDC table, can we simply select "Resume" to continue processing where the task left off, or would we need to resume via timestamp?
My main concern here is to avoid missing changes. Help with this is appreciated. Thank you.
Kind regards,
Mohammed
Hi @MoeE ,
Regarding task execution,
"While the task is running, one of the MS-CDC tables in the source database is deleted, and the QR task fails and stops."
By default, the expected behavior should be that Qlik Replicate drops the corresponding target table and continues running.
When you drop a CDC-enabled source table in SQL Server, its corresponding change table is also removed, which means all captured changes are permanently lost. As a result, resuming replication from a specific timestamp is not possible once the change table has been deleted.
Regards,
Desmond
Hi @DesmondWOO,
Thanks for the response. Theoretically speaking, if we archived the CDC table on the source, deleted the original, then restored it, how would Replicate behave? Would it simply suspend one table or would the entire task stop? If unknown, I will try to setup this test in our environment.
Regards,
Mohammed