Re: REUSEDLT(*NO) vs. REUSEDLT(*YES)

Or just let the DB component keep using the deleted records until it all settles out.

Yeah, I was aware of such a situation - chose not to mention it - giving a very general answer, I suppose.

So I have my "could" - you have your "occassionally" - seems we're on the same ground here!

Cheers
Vern

----- Original Message -----

On 9/10/2013 2:03 PM, Vern Hamberg wrote:
I see no problems with this - seems to me, that you could end up with a smaller file - only active records, for the most part, no wasted space for deleted records.

Hardly. Suppose an application program _inserts_ (or writes) 100,000
rows (records) into this table (or file), then later, another
application program _deletes_ 50,000 (not necessarily contiguous)
records. There will now be "wasted space" for 50,000 rows or records,
regardless of the setting of REUSEDLT *YES or *NO. So, you could still
need to use RGZPFM occasionally if you really want to get file sizes
down to the "minimum". /*
*/

Just curious, do you have something in mind that might cause problems?

Vern