Reconcile hangs for want of data and there is no test table to force transactions on source datab (4330949)

If running a reconcile, the process sp_oreconcile will wait for user activity on source to push a message in the queue that is from a redolog higher than the one up to which the queue is reconciled. Till it happens, the sp_ctrl may appear to be in a hung state. This may add to the maintenance time.

Reconcile hung for want of data

Normally one can wait for user activity on source database for reconcile to receive the desired message from next redolog and finish normally. However, at times this may involve long wait. If there are test tables in replication, one can always force a transaction on source database. If there are none, then as a workaround, one can issue a flush from source sp_ctrl as follows:

sp_ctrl>flush o.SID

The flush will generate a message in Shareplex queues (due to Shareplex writing in redolog) that will make reconcile exit so that other maintenance activities can be performed.