With what error did the switch job terminate? Were there any dumps in transaction ST22 at the time the switch procedure terminated?
Was there sufficient space available for the messages to be copied to the active container? When the switch procedure starts, messages are copied from the SXMSPMAST to SXMSPMAST2 or vice versa, only when the switch procedure is finished is the inactive container dropped, the table from which messages were copied.
By using the parameter MOVE_INSTEAD_OF_COPY this problem can be overcome, messages are deleted as soon as they are copied across rather than waiting for the procedure to complete, therefore saving on DB space consumption.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.