Re: A Mystery SQL Problem

[Booth Martin <booth@xxxxxxxxxxxx>]

One step I take in a situation like this is that I purposely do 
something I know will compile-but-break the process.  Quite often in 
looking at the broken pieces the  culprit spits in my face. For instance 
I might copy both files to a sandbox.  If its a big file, take the last 
thousand records, or such.  Then perhaps yse an existing field to update 
with an incremental counter and then watch the counter for odd behavior.


On 12/6/2016 4:41 PM, Marvin Radding wrote:
> This program is use to build records for the AHPOVRP file. When exiting the program, the records, identified by the effective date, are copied to the MMNYOVRP file.  When they do this on a state by state basis, the duplicates are always at the end and this will move as more states are built. For example, when they build the first state, it duplicated 40 records. When they built the second state the duplicates for the first state disappeared and the second state had duplicates. And when they built the third state there were no duplicates on the first two states but then the third state had duplicates.  Each time, the program will extract from the AHPOVRP file all the records built up to that time and will only crate duplicates for the last few records. I am very mystified.
>
> Thanks,
>
> Marvin
> No trees were killed in the sending of this message, but a large number of electrons were terribly inconvenienced