March 4, 2013 at 7:05 pm
So I have created a new Tabular cube on 2012, deployed it for a while. And created a few new partitions via SSMS in the server to cover all the data.
Now I have made some changes in the project, planning to deploy it to the server, but can NOT find an option to ignore existing partitions (that are not defined in the project itself)
I have tried it on a test db, and no matter what i do... the project seems to deploy its definition, and overwrite what's on the server. meaning all the partitions that i created after deploy are WIPED!
I have billions of rows of data, so reprocessing all the missing partitions are really not preferable... and managing the partitions in project but not in ssms is also not preferable as we typicall use scripts to add / manage partitions after it goes live
we have found some blog posts about changing .deploymentoptions file... but we don't know what value we should change to (not in BOL).. we changed it to "RetainPartitions" as a test.. but that doesn't work
anyone has been through this? thanks
May 22, 2013 at 11:16 am
I'm having the same issue. If you have figured out a workaround, please post it. As of now I think I'm going to create partitions based on querys that returns no data (where 0 = 1). Then I can switch that after a deployment.
August 1, 2013 at 5:02 pm
just wanna loop back to the original post.. and say that thanks for the suggestion to BISM normalizer.. it definitely works.. and it's actually a much simpler deployment process now than the right click deploy from SSDT (which is always scary)
thanks!
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply