All,
Just more of a post to build awareness and methodologies that the community can ponder on, contribute to, or simply ignore :)
This past month we have been preparing an upgrade from our 2012 CU3 installation up to FP1 + CU5 prior to a rollout of R2. We found out several interesting facts and came up with an approach that worked for us. I'd be interested to know if there are more efficient or expedient ways of going about this process.
We found CU5 upgrade alone placed modifications in the SYP and FPP upgrade layers, and had a model display name containing the value "KB 2828929". We found FP1 upgrade alone placed modifications in the FPK and FPP layers, and had a model display name of either "Extensions" or "Extensions Upgrade". We found that when installing FP1 and CU5 simultaneously via slipstream, there were models with "Extensions [KB 2828929]", meaning they were modified by both Feature Pack and the cumulative update.
Our primary concern and objective was to find all objects which we had modified in the CUS layer, which were also modified by either CU5 or FP1.
To accomplish this, I created a Project in the AOT and used the Advanced Filter to populate it with objects in any of the identified models from above. I then created another project with any objects modified in the CUS layer. We then exported these lists of objects out to Excel and performed duplicate value analysis. We then found our list of impacted objects, and have reviewed each one in a final project that we will be checking in to TFS for deployment to QA and eventually Production environments.
This worked well for us, although a bit tedious. I have the feeling that these kinds of large updates simply demand acute attention to detail, however if you have any tips or tricks to make the process more expedient please do tell!
Cheers,
~S