Re: Trigger programme without a trigger.

Wilt, Charles wrote:

Ding, Ding, Ding....

To your corners gentlemen...

<smirk> Ah, me and Rob just like to dust it up a bit, is all...

Joe, it is possible to write a faulty trigger program just like anything else. Perhaps it was simply
a missed business rule, the trigger should have caught the problem but the specs were incomplete
<grin>

Regardless, the CFO says fix it, and Mr. QSECOFR does just that with DFU or SQL. With the trigger any
business rules, except obviously the one that was missed, would still be enforced. With an I/O
module, they wouldn't.

Okay, I'm going to give you guys this: the trigger can perhaps apply some of the business rules to DFU or SQL changes. My point on this is twofold: it doesn't catch them all (if it did, you wouldn't be doing DFU in the first place), and second it perpetuates the idea that DFU is a "necessary evil" and that it is acceptable to run DFU/SQL on production data.

I'm sorry, but I don't believe that. Any time DFU is required on production data, red flags ought to be gonig up all over the place - if you were a SOX shop, it would take longer to do the paperwork than the change!

Joe