We have a deployment scenario where one environment is expected to have a large number of machines, potentially hundreds.
The machines will share application code.
Unfortunately, due to issues outside our control, most of the machines will require different connection strings.
I would prefer not to keep such a large list of variables in Octopus Deploy as I think it will be a maintenance headache.
I would prefer to store the list of connection strings in a file that we can keep in source control.
My current plan is to have a file that lists all the machines and their connection strings and to process this file in a post build step to generate a configuration transformation file per machine that can be run using Octopus Deploy’s additional transforms capability.
Unfortunately this will result in rather a large number of files, so I was wondering if anyone had any other suggestions.