Hi, hope anyone can help me.
i have 3 databases and 1 should hold backups from the other 2 DB's. To do this I have written a scheduler to start a task, which executes a procedure.
The procedure copies the data via dblink in following format:
Insert into table tablename select (...) from table@dblink!
For one Database this works fine. The Task runs when no users are working normally at 20:00 pm.
But the other fails with different behaviors. Mostly an 2063 occur after 56 from 58 tables (often at empty tables).
Sometimes a 3113 or 3114 occurs.
The thing is that if ora 2063 occurs, no errormessage from the gateway is included so that i not know what the problem is.
I have tried different things like monitoring the router and network traffic while copying, creating a large RollBackSegment for this Transaction (Ca. 500.000 lines before commiting the Transaction), checking the Tracefiles on the Remote Machine.
When I comment some tables with a large amount of lines sometimes the procedure succeed, but not ever.
In several cases the task crashes without error and I have to restart the programm, but the problem in this case is that the last accessed table is locked and by copying again a get: try whith NOWAIT - Resource is busy (I have forgotten the Nr.)
Is there anybody who can Help identifying or fixing the problem?
Thanks
i have 3 databases and 1 should hold backups from the other 2 DB's. To do this I have written a scheduler to start a task, which executes a procedure.
The procedure copies the data via dblink in following format:
Insert into table tablename select (...) from table@dblink!
For one Database this works fine. The Task runs when no users are working normally at 20:00 pm.
But the other fails with different behaviors. Mostly an 2063 occur after 56 from 58 tables (often at empty tables).
Sometimes a 3113 or 3114 occurs.
The thing is that if ora 2063 occurs, no errormessage from the gateway is included so that i not know what the problem is.
I have tried different things like monitoring the router and network traffic while copying, creating a large RollBackSegment for this Transaction (Ca. 500.000 lines before commiting the Transaction), checking the Tracefiles on the Remote Machine.
When I comment some tables with a large amount of lines sometimes the procedure succeed, but not ever.
In several cases the task crashes without error and I have to restart the programm, but the problem in this case is that the last accessed table is locked and by copying again a get: try whith NOWAIT - Resource is busy (I have forgotten the Nr.)
Is there anybody who can Help identifying or fixing the problem?
Thanks