dschlieder
Well Known Member
Been searching here and on the KG and can not find any reference to my issue.
A few weeks (or months) ago, we noticed that our about half of our COA records had a last update date of 0.
So we picked 1/1/1999 as a date and updated them (almost 500k records).
I tried then to duplicate the issue based on the PID in those records and couldn't. Monitored for a few days and no issue so I stopped monitoring it.
Today I've found another 8k with the last updated of 0 again!
I've spent that last couple of hours going over C code, programs, UBEs that touch this table and have not been able to find anywhere where this can happen.
Anyone seen this before and know of a resolution?
The reason this is important is the ETL for COGNOS ends up processing all 800k records when any one has a last updated date of 0, even if there isn't a change the it cares about.
Thanks for any pointers..
A few weeks (or months) ago, we noticed that our about half of our COA records had a last update date of 0.
So we picked 1/1/1999 as a date and updated them (almost 500k records).
I tried then to duplicate the issue based on the PID in those records and couldn't. Monitored for a few days and no issue so I stopped monitoring it.
Today I've found another 8k with the last updated of 0 again!
I've spent that last couple of hours going over C code, programs, UBEs that touch this table and have not been able to find anywhere where this can happen.
Anyone seen this before and know of a resolution?
The reason this is important is the ETL for COGNOS ends up processing all 800k records when any one has a last updated date of 0, even if there isn't a change the it cares about.
Thanks for any pointers..
