Re: Phantom Inserts

In article <1995Feb27.145749.18654@dbsoftware.com>, janco@dbsoftware.com (Jim Ancona) writes... . . . >How does HOLDLOCK help you in this situation? The first SELECT with HOLDLOCK keeps a >shared lock on the rows in the SELECT for the duration of the transaction. That will >prevent another transaction from updating or deleting those rows, but I don't see >how it will prevent the INSERT from showing up in the second SELECT. Am I missing >something? >-- >Jim Ancona janco@dbsoftware.com jpa@iii.net >Opinions expressed are my own, and not those of D&B Software. > I think it was stated in here somewhere that it will also hold the page locks for the index buckets meeting the criteria. This will block another transaction from inserting into the index a tuple with the same value. -Jay Feenan ********************************************************************* *** Jay Feenan *** *** Oracle Rdb Engineering *** *** *** *** opinions expressed here are solely my own and not those of my *** *** employer. *** *** *** *********************************************************************