Jim writes back: >>> Therm-O-Link <[log in to unmask]> 07/02/98 09:33am >>> Mike Berkowitz writes: >DBPUTs don't care if they use the delete chain, or the high water mark to >determine record placement. However, if you don't repack your data set >after your mass delete, any program that serially reads the data set will >continue to read to the high water mark, which could be well past the >last record. Do you mean "repack your data set" as in using Adager/Dbgeneral to reorganize the data set (which I can't do in a 24x7 environment)? Or "repack your data set" as in delete all entries while writing them to a flat file, sort the flat file, and then add the records back in (which I can do, but I don't think it will reset the high water mark)? Jim Phillips Manager of Information Systems E-Mail: [log in to unmask] Therm-O-Link, Inc. Phone: (330) 527-2124 P. O. Box 285 Fax: (330) 527-2123 Garrettsville, Ohio 44231 ------------------------------------------------------------------------------------ You can use either "repack" you mentioned. If you delete all entries, the highwater mark and deletechaincount are both set back to zero. Of course your applications may have a cow about the temporarily missing data. Mike Berkowitz Guess? Inc.