Originally Posted by
golfbananajam
If I understand it correctly, the data provided to NATS has been pre-processed and so any errors (mistakes) should have been caught by the system doing the pre-processing. If my understanding is correct, I would expect that the assumption is that the data is now ONLY valid data, so this failure case isn't as straight forward as you may believe.
I work in the work of complex software, testing is a nightmare. As others have asked, at what point do you stop? How many bugs do you release (bearing in mind you may not know about them)? These are questions we're constantly asking and we never ever get to a stage where we satisfy everyone.
This answer (aka 'excuse') has been repeated so often, ie 'testing's soooo complicated, we couldn't possibly capture every permutation (which actually means not process an incorrectly formatted message). Really??
'at what point do you stop?'. Er, when you've built one fall back, and preferably two - if, as you say, this is really a mission critical system.