Half and half?
Looks like the back end is churning away doing all the right things, and the front end is totally fucked.
Probably soon after the problems started someone decided that the bank was up and running and handling the usual business fine, and all that was needed was to fix the web front end and all would peachy. So forward, no rollback, we have migrated successfully, break out the champers and put up a notice apologizing to those few users who were having online problems (temporarily).
Just a bit of Agile fail early fail often with those weird webby types. Couple of sprints and all fixed.
By the time it was obvious that online access was borked long term and wasn't going to be sorted any time soon a couple of days had passed and reverting to the old system was nigh on impossible and getting harder every second.
The only option appeared to be to plough on and pray for a miracle.
If there is not enough hardware to handle the front end load then it's going to take more than a quick trip to PC World to stand up more kit and get it integrated. This may be where IBM come in.
Biting the hand that feeds IT
