Schema won't let me update due to a join to a dataset that doesn't exist.

[DanSpigs1]

I am trying to make an update to a schema in our Dev environment, however when I hit save it gives me an error

"This operation to update mixins: *CLIENT NAME* Contact Details is not possible as it is used by schema: *CLIENT NAME* Core Customer Profile which participates in a union operation: https://ns.adobe.com/xdm/context/profile__union which is referenced by Dataset: [67ec132ff008ba2aeec69197, 67f7de8e8dd71e2aef9a20f2]"

However, I have deleted out those datasets.  Is this just a timing issue that it takes time for the system to recognize the datasets are deleted or do I also need to go delete the dataflows?  All dataflows were one-time executions so nothing should be running there.

Is there a way in the tool to get the actual names of the datasets being referenced?  My only thought at this point is it's an OOTB dataset, in which case, how would I go about making changes to the schema?  Wouldn't it be locked in forever?

[supratim320 · Accepted solution]

Hi @DanSpigs1,

Are you trying to delete or add a few fields from that field group (mixin)?

If delete yes: then my assumption of the error: I think those fields are already part of some schema (and hence the dataset) and they were enabled for profile. So, now although you were able to delete the dataset, those fields are already part of (union) profile schema, hence the deletion of field are not possible.

Q2: "Is there a way in the tool to get the actual names of the datasets being referenced?" - if you have the id of the dataset, I have seen AI assistant can be helpful here.

Hope this helps!

[maggod]

Had a similar issue recently. Turned out that the mentioned datasets are the SNAPSHOTs. Can you check if the ids do match? If yes, there is no way to update the schema except follow the schema evolution rules.

[Supriya_Savre]

Hi @DanSpigs1,

Did the shared solution help you out? Please let us know if you need more information. Otherwise kindly consider marking the most suitable answer as ‘correct’.

If you've discovered a solution yourself, we would appreciate it if you could share it with the community.