changes to adf database linked service not changing via devops to prod

Question

I have my ADF V2 Development environment connected to devops and am successfully creating a branch, making a change, making a pull request, publishing and releasing into production - also using the MS developed PS script to stop and start triggers and delete deleted pipelines.
Following this normal CI/CD process I recently made a change to the DB Connection linked service - standard Azure SQL database connection.
Previously I was using the manually entered fully qualified domain name like devdb.database.windows.net with the database name, sql authentication of user id with password.
The change I made to the DB linked service was to create a parameter for the server name and use this in the domain name so it is now @{linkedService().Server}, I also got the password into a Key Vault and successfully connected this all up in my feature branch. Merged into my master branch and published. It is working fine in the Development environment.
But when I do the release into Production these changes are not being made to the database connection linked service. I can't see where these values are captured in any of the Template files - are changes to existing linked services able to be migrated via devops?

Thanks

Answer 1

It would seem to me that making a change to the DB Linked service and then following the DevOps process of merge, publish then deploy to prod may not be the best way. To resolve this I created a new branch, deleted this db connxion then created a new connxion, assigned the new connection to the datasets ( which wasn't many) then saved, created a pull request and merge back into the master branch. Then published to adf_publish and did a release. Had to edit the release and go into the global parameters and delete the old dbconnxion line, the newly named dbconnixons was present. Released and it worked fine in Prod.
Probably lesson learned is to have your Linked Services correctly set before initially going to prod otherwise it will be a pain to fix if you have a lot of datasets to edit.