Posted by: Martha Ho in 12.1 on
Jun 4, 2009
Today's dilemma...Everyone wants to know where the project stands! My delay in blogging is purely attributed to trying to decide how to proceed. A simple thing, but not always as simple with regards to upgrades when serious issues surface on a new release.
Testing went well; individuals were able to get thru all their test scenarios, updating them as they go. Good news, we don't think we will uncover too many new issues moving forward. Bad news, they did uncover major issues that our day to day end users will deem to be show stoppers. They have to be resolved by scheduled go live.
We were hoping to go live end of June. An internal target, but still a target that the upgrade team is being held to. We internally have to figure out which particular days given month end, quarter end, etc.
Back to the dilemma...do you move the go live date now or plow all the way to the end and potentially wait.
Entire testing team was brought together, vote occurred - Plow ahead.
We'll continue to request Oracle on a daily basis for their fixes/solutions/workarounds and appropriately escalate to minimize impact on schedules. Oracle is also going thru discovery and providing fixes that will be new, so potentially the iterations of testing and retesting will be there, so we'll need to keep our current testing instance. We'll then request permission from our Steering Committee to move forward to another instance and document the known issues. This means that our own internal Oracle production systems will be left with minimal instances to resolve issues should they come up, but worth the risk given minimal issues we do have day to day.
We will then upgrade the next iteration, Iteration 2, which will be our user acceptance instance. By performing the upgrade on Iteration 2, our Oracle application DBA's will be able to predict better timings for the final production upgrade, currently they need 1 week to complete; hoping for a better number at the end of Iteration 2, no one can be out for that long. Again, the first iteration is always discovery for all teams including DBA's. They too, had to work thru issues with Oracle.
The internal goal for the team is that we'll work with Oracle come up with either fixes or workarounds that will not affect the upgrade scripts the DBA's will be running in Iteration 2 and that we'll be able to simply patch at the end.
We can then perform user acceptance testing (UAT) between our global teams and keep to our schedule.
