This likely happened because you had 2 accounts on the same server. Then, one was migrated by them, creating the problem we're facing.
Once the second account, whichever it was, migrated too, you were good.
The problem is that we can't speed up by, for example, creating new accounts to replace old ones. It's like they intentionally create new accounts on the old server to prevent us from creating new accounts and moving our data ourselves if this feature is important enough, using more of their resources.
They're very smart people. It's too much of a coincidence things are broken in a way we can't work around, almost as if it were intentional. Otherwise, if you're migrating people to a new datacenter, why in hell would they put NEW accounts on OLD servers just to migrate them later?
They possibly want to break things for heavy users that would move their data manually if that were possible.
It feels they're intentionally breaking it to "quiet fire" unwanted customers.
If you need the feature you can't just sit around for 9 months on a gamble to know if or when you'll leave.
Can someone explain why would it make sense to have NEW accounts using the OLD server if you're migrating everybody? Or are the load balancing between two datacenters and there will be always a group in one side, another in another side, and the problem will never be solved and we can be always relocated without warning based on hidden criteria like usage, behavior, geo location etc?
Sorry for ranting.