Remarks to FUAM - GT-Analytics/fuam-basic GitHub Wiki

Important to know

  • There is a maximum of 500 requests a 100 workspaces possible through the scanner API (learn.microsoft.com). If you have more than 50.000 workspaces in your organisation, which are attached to capacities, the pipeline must wait one hour for the next Scanner API call. This might lead to a longer Pipeline run.

  • There can be errors in case specific item types have not been created on the tenant, yet. We tried to reduce these kind of errors, by catching this kind of errors, but on relatively empty tenants this could still effect the execution

  • In some cases the reports throw an error because of missing fields, which have not been provided by the API. In this case please execute the following steps:

    • Try to refresh the underlying semantic model. Check if there is an error as in the following picture: image

    • Open the semantic model image

    • Click on "Edit tables" image

    • Press "Confirm" to refresh semantic model meta data image

    • Test semantic model refresh & report

Known Errors

  • There are some known issues on "empty" or demo tenants, where some objects do not exist, which causes errors:
    • If there is no workspace description on the whole tenant. In this case just add one workspace description. This will fix the error
    • In case there are no regular scheduled refreshes on the tenant, the execution for capacity refreshables can fail. This should be resolved by creating a scheduled refresh and running it multiple times
    • In case the are no delegated tenant settings set in one of the capacities, the extraction step will fail. You can remove this step if it is not needed in your tenant
    • FIXED (see Initial Deployment Step 6.): Currently the notebook within the pipeline "Load_Inventory_E2E" is using the notebook owners identity to query the metascanner api. In case the user doesn't have permission, this will fail. We are working on an alternative solution