What's the best path for Cloud > Server migration (ps we use Zephyr!)

[Steve A]

For some reason Jira Cloud > Jira Server seems to be a Lunar-Mission-level-effort for Atlassian. Here's what we've tried and the results:

1. Backup (using backup for server) and import.
   • This went OK except that it over-wrote all settings and tried to install our cloud plugins on server, which were incompatible and seemingly impossible to uninstall. We had to roll back our server & database after this.
2. Export all issues from cloud and bulk import them into the server project.
   • This required us to manually select the 100+ issue types and/or create fields if they didn't already exist on the server project.
   • It also didn't import anything for Zephyr.
   • It also imported junk files for all the attachments, which were oddly named the same way but had html in them instead of the original content, which printed an error message. I was actually impressed by this one, it would have made me laugh if the content were subbed out with, say... Zombo.com. 

Things are still 'up in the air' and would appreciate any advice on landing this bird.

[Alexis Robert]

Hi @Steve A , 

there is a pretty good documentation on Atlassian site for the Cloud to Server migration : https://confluence.atlassian.com/confeval/other-atlassian-evaluator-resources/migrating-from-cloud-to-server

There are some specific steps to migrate attachments as you will see. For Zephyr, I believe it needs to be done separately with the internal import/export feature.

Let me know if this helps, 

--Alexis