how to start one of the tables with "Advanced Run Options"
I have a 10-table replication task to Kafka, and need to force one table "Chane Processing" from a point in time. But "Advanced Run Options" seems only working for the entire task. Reloading that single table is not an option either because the target is kafka - if reload, there will be duplicated data.
Not sure what's the source type and why the single table should be 'replicated faster' than other tables. is a dedicated task (with this table only) meet the 'real time replication' requirement? certainly it may add additional load to source side.
Regards,
John.
Help users find answers! Do not forget to mark a solution that worked for you! If already marked, give it a thumbs up!
The add prefix can be added to any load, select or map...using statement in the script.
It is only meaningful during partial reloads. During a partial reload the QlikView table,for which a table name is generated by the add load/add select statement (provided such a table exists),will be appended with the result of the add load/add select statement. No check for duplicates is performed.
Add Only :
If you add the only option to your statement ex: ADD ONLY then the partial reload will not be used when run the normal reload.
That table has not been replicated for a couple of days so I want to force its replication from when it's stopped. The source is a SQL Server Pord db. (it may sound weird but I don't have access to the DB).
>> That table has not been replicated for a couple of days
Why/how is that? Was the underlying problem fixed? Is it currently loading changes for that table? From what time?
>> The source is a SQL Server Pord db. (it may sound weird but I don't have access to the DB).
That is weird, as to effectively support Replicate it is really very desirable to have (source) database access.
Think about how Replicate works. It has a single source reader thread, going over all the changes logs by time, from archives if needed up to the head of the active log. So your main task is not close to, or at the head of the active log time'X'. You want it to 'stop' go back in time, start reading old (archive) logs, skip non-selected events, look for this table you want to refresh only and then when it is back at 'X' which may take hours, resume looking for changes from all selected tables. That's not in the current code, unlikely to ever be added.
The proper solution today is to create a temporary task, as @john_wang mentions. I fear however that this will also fail because you may not be able to get all the required change logs readily accessible. Try it. You may need a filter to stop reading changes already picked up recently, and then stop the helper task once caught up.
Are there time-and-date columns on the rows to be loaded? If so, a possible alternative, is to create a helper full-load task just for that table and add a date-and-time filter based on some 'last insert' or 'last update' column which many applications have.
@anat - I think you are referring to QlikView, not Replicate as the OP asks for.
