SCA changes.html identifies several objects that are not in the script

codegumbo

We're using the Powershell scripts to deploy a Nuget package from SCA, and the script tab shows only 5 objects (expected) to be deployed, but the list of changes shows nearly every object as being added or modified (114).  Why?  Where should I look to start troubleshooting this discrepancy?

GeekyCat

Did you find an answer on this?  I had the same thing happen and it re-deployed all the additional objects as well

Kendra_Little

Hello,

Apologies that @codegumbo 's question back in September was missed. We have some automation in place to try to help us make sure we get to these questions but it looks like something went awry in this case -- we're checking into that separately.

I suspect that in this case the projects are authored in a SQL Change Automation Project, so I'm going to answer that way, but please let me know if you're using SQL Source Control.

The first thing that I would look at is the Drift report.  Are there objects in the target database which have diverged from the code in the project?

The Changes report is generated by comparing a snapshot of the expected schema at the end of deploying your project (which is captured during a build). If you've had drift occur then that comparison may show things which are unexpected -- the Drift report can help explain it. 

Kendra

codegumbo

Unfortunately, that was not the case with us.  I sent over both the change report and the drift report, and support couldn't spot anything either.

Luckily, it seems to have been a 1 time occurrence for us (as frustrating as that is from a technical support perspective).  We haven't had an issue with this since then.

GeekyCat

Hi Kendra - Thank you for your response.  I am using SQL Source Control.  I'm sorry I didn't provide that first.  Between Dev and QA, there are objects that I don't want synced and I've created a filter from SQL Compare to use a scpf file which is in the a folder that the agent can access.  

I have a changeset that, when I look into it, I see the 2 scripts that I committed associated with the Release Pipeline. (Image 403).  I've provided screenshots that hopefully will provide the needed information.

But when I redeploy the stage object, you can see in the sample logs that it's actually doing so much more.  Including dropping tables that have actual data in them.

Since QA (and Prod) contain update and refresh scripts for those environments, I can't make each environment exactly the same.  I'm sure I'm not the first person to do things this way.  Is there a better way of setting things up?

Please let me know if you need more information and thank you so much for your help

Cathryn