During my recent workshop at client aimed at working though bank rec issues and performance items we found that the current monitoring we have is not sufficient. We had to extend this with the use of a trigger. Afterwards we could easily locate problems. The code for this trigger and extra columns is attached for reference
I think we should extend this properly into the product.
Client have also requested this because they will be able to create reports that show degradation tends of highlight troublesome accounts
Proposed changes
alter table bank_import_monitoring add automatch_server varCHAR(254);
alter table bank_import_monitoring add intraday int;
alter table bank_import_monitoring add automatch_datetime_start datetime;
This allows us the ability to see the time it takes to match individual accounts, and more importantly the time the automatch started, and the delay between statement load time and automatch
The code would also need to be changed (rather than the trigger)
So that the 3 columns are populated when the automatch begins
If the automatch processed one account, which means that it did not have to automatch other items in the queue for that same bank account (this happens). Then rather than just setting the automatch_status in bank_import_monitoring to 'PROCESSED' can we set it to something else so that this is clear like 'MATCHED BY XXX' so we know another match actually ran superseded it.
I can provide a client database and bank import files and setup if required, however I'm hoping we can just make this change using standard NUNIT type databases