Then each TR was progressed up the tiers, N1dev tr's stopped at N1QA while its associated break/fix change went into production.ĭuring the near end of the project (when we hit QA i think) we froze TRs in N landscape unless they were critical business impact. Every single transport that was released from NDev was first imported into N1Dev, a TOC was created for it and notated. We manually tracked the N-TR and its associated N1-TR counterpart.

We then used Transport of Copies of new work from NDev to N+1Dev, a new TR was copied off the NDev TR and/or changes were manually made per the EHP requirements. Transport Dual Maintenance Once the landscape was built, we performed dual maintenance of transports from N to N+1 landscape. We then built out new individual systems and then performed clones from Prod -> N1xx. Landscape buildout N+1 was our EHP path, prior to the buildout of N+1 we did a remote client copy from Prod/800 -> NDev/300 (our unit testing client). We built out an N+1 landscape to accomplish our upgrade. When doing the cutover do several backups in different ways, copy to disk, copy to tape, dumps, etc.Īnyways. They ran several reports like EW but they're called go-live check, they helped with the sizing of the system. Involve SAP, let them know what you're doing, talk to your client rep and see how they can help. Test! a lot! Even the upgrade test it, do dry runs.Ĭheck the SUM logs and see if anything can be optimized, like adding an index or parallel processing. Also take into consideration that if you have an SAP BW I believe it needs to be 7.4 in order for it to work with EHP7. Keep in mind that if you follow this approach your dev team will need to freeze some projects during the upgrade or code in both landscapes. IDK what size your prod is but having a dev system with a big database at least in our case is a no no, you'd be wasting resources. That was our 'main' landscape for the project, one dev and one prod (which later bacame a QA).

we cloned dev and upgraded it and cloned prod and upgraded it. Pretty much what we did was clone our landscape, anything and everything that interacted with our ECC got cloned. I don't know what kind of hardware or infrastructure you have but we have virtualized systems so that helped a lot. A couple of years ago we did an upgrade from ECC 5 to ECC 6 EHP7.

Ecc 6 ehp 7