For one project, I setup ODK central on AWS and collected data in phase 1, which is about 35GB. Now for phase 2, the organization wants to run the data collection on their own hosted physical server, and the data needs to be migrated to that as well.
How to do it properly?
Since the new server is not on AWS, I cannot take snapshot and deploy it there.
Its not compulsory that I take everything strictly as-it-is to the new server (including CENTRAL config). I can setup a new central software on the new server and migrate the data from old to new machine (how?).
Yes, there are ways you may migrate the complete VM instance from the cloud to the local infrastructure. This documentation explained the process other way around; however, it can be done from AWS to local server using the export import tool.
So although on the new server I need to deploy a newer version of the form (against which i collected old data), I should probably first deploy old form version on the new server, upload previous data, then deploy upgraded version of the form, and then start new data collection. Right?
Just sharing simple steps for the benefit of others who need guidance on similar matter here.
Looking to save time? Try ODK Cloud, the official managed hosting service from the creators of ODK.