The issue now seems to be solved. Migrating a snapshot without OID provider seemed to do the trick together with creating an empty administrator role to inherit BIAdminrole (2024 admin role) as BI Service Administrator did not seem sufficient. There were no additional configs that we had to do, other than what was in…

Thank you for your response! We have not found this option. The one that most closely resembled it is "Client As User Principal Enabled" - Is this the same?

We have now attempted to both set ownership of all objects to an IAM-adminuser2 that does not have a corresponding OID-USER in content manager and then export the snapshot. This yields the same results. IAM-adminuser2 has all privileges, where adminuser1 lacks some of them. Another attempt was deleting the OID provider…

Yes, it is via group membership. Nothing is given directly (permissions/roles). We will definitely check the solution you have outlined, Thank you! We're currently in the process of attempting to change all object ownerships from source environment to an IAM user and then migrating the snapshot to see if that works (tests…

I was afraid of this. While we have Oracle database, we'd like to avoid direct queries to other systems from it, so we will likely build another solution instead. This helped a ton as always, thank you!

Thank you!

@Gianni Ceresa, this is true. I also left a SR to Oracle for them to answer, which came back ambiguous - "Should be L7.9 or 8". If someone are able to check with their marketplace version, it is greatly appreciated! Although it would be nice if Oracle knew this too.

It turns out that it was not that important to get them in the same dataset after all, as it will only indicate a possible error source. My solution was neither elegant, nor perfect yet still good enough. Thank you for your contribution to my thinking, Gianni!

This solved it for me, it seems that there was indeed something "wrong" about my year attribute, so I exported the historical data, and the viz worked right away when I uploaded it. Thanks to all contributors, I am always amazed on how quickly my problems are solved.

The data type for år (year) is Date (unless there are specific date datatypes that I am not using), which I am using. As linked in my pictures I am able to choose the forecasting option, there just seem to not appear, which is why I am a bit confused.

That is what I am attempting to do. The one with the grey background is the one with forecast turned on, but it looks the same (and the numbers are also the same) - I had expected there to be a 3 more years on the x-axis, a confidence interval and a line for 3 more years based on the attribute year (år) and measure amount…

Ok, then I conclude that I am not completely stupid, thank you! I've gone ahead and left an enhancement request to suggest an "improvement".

While I do understand that, it is unclear for the user when selecting standard/default. If I choose default settings in any system, I expect it to look up a config file of some sort, not what I myself as a user chose in a previous selection.

When I look it up in DV: 'date' in I use in dashboard prompt is datatype: Time, is set to be treated as attribute and the table that it presents in a workbook does not show all dates (saturdays and sundays are excluded - I anticipate that the db has no records on saturdays/sundays), but has single entries per date - for…

When seeding with a specific date, it does not hit the cache, so I am a step further - thank you!

No, it is not the seeding that takes a lot of time (it seems to run fast) - it is the dashboard loading. After seeding the dashboard uses a lot of time to load when seeded with distinct value, while when seeded with server variables (both values being '01-01-2023 00:00;00') results in vastly different load times. Seeded…