All ideas are welcome. Just because the Idea doesn't make it into the product immediately does not make it a bad idea.
As we start refreshing tables over and over, sometimes certain tables will fail silently, or something problem in legacy will occur such that the Data Refresh is not complete and there is no way to know this.
Can we build a Alert system (active or passive) where we can identify those Tables in Collect where the refresh is materially different (+/- 10%, or whatever) from the last time the table was refreshed. Understand that this won't be exact, but would be a good mechanism to identify potential problems BEFORE the team starts processing all the objects and only then discovers that one of the legacy tables failed, and now it's difficult to refresh that table because it's been week since the rest of the data was refreshed.
At 3M this was common with our CMR data refeshes from Oracle On Demand (cloud system). The refresh process was inconsistent, and often a few random tables failed to refresh. At KHov, I've had several example where one JDE table fails to refresh but gives no error. F0005 has 100,000 records every time, but suddenly returned 0 records for no apparent reason. Refreshing again returned all the expected records. Tough to catch if you are refreshing 1000 tables and only a few random tables are different.
Guest
