Jaa


FinOps toolkit troubleshooting guide

This article describes how to validate FinOps toolkit solutions were deployed and configured correctly. If you have a specific error code, review common errors for details and mitigation steps. If you need a more thorough walkthrough to validate your configuration, use the following steps that apply to you.


Do you have a specific error code?

If you have a specific error code, we recommend starting with common errors for a direct explanation of the issue you're facing. There's also information about how to mitigate or work around the issue.


Validate your FinOps hub deployment

Use the following steps to validate your FinOps hub deployment:

Step 1: Verify Cost Management export

  1. Go to Cost Management exports and make sure the export status is Successful.
  2. If it isn't successful, ensure you have the Cost Management resource provider registered for the subscription where your hub is deployed.

Step 2: Verify Data Factory pipelines

  1. Go to Data Factory studio, then go to Monitor and make sure both pipelines are running.
  2. Compare the last run time with the time of the last cost export. They should be close.
  3. Open the Data Factory instance in Data Factory Studio and select Manage > Author > Triggers. Verify the msexports_FileAdded trigger is started. If not, start it.
  4. If the trigger fails to start with a “resource provider isn't registered” error, open the subscription in the Azure portal, then select Settings > Resource providers, select the Microsoft.EventGrid row, then select Register. Registration might take a few minutes.
  5. After registration completes, start the msexports_FileAdded trigger again.
  6. After the trigger is started, rerun all connected Cost Management exports. Data should be fully ingested within 10-20 minutes.
  7. If the ingestion pipeline isn't running and it's showing a MappingColumnNameNotFoundInSourceFile error message, verify the export is configured for FOCUS 1.0-preview(v1) and not 1.0.

Step 3: Verify storage account – msexports container

  1. The msexports container is where the Cost Management pushes "raw" export to. This container shouldn't have CSV files as hubs transforms them into parquet files.
  2. If you see CSV files in the msexports container, refer back to Verify Data Factory pipelines.

Step 4: Verify storage account – ingestion container

  1. The ingestion container is where clients, like Power BI, connect to pull data. This container should always have one or more parquet files for each month.
  2. If you don't see any parquet files in the ingestion container, check for CSV files in the mseports container.
  3. If you find CSV files inside the msexports container, it means that Data Factory pipeline isn't working. Refer back to Verify Data Factory pipelines.
  4. If there are no CSV files in the msexports container and no parquet files inside the ingestion container, it means the Cost Management export isn't running properly. Refer back to Verify Cost Management export.

Validate your Power BI configuration

Use the following steps to validate your Power BI configuration:

Step 1: Identify your storage URL

Before you begin validating your Power BI configuration, you need to know whether you're connecting to your data using one of the following mechanisms:

  • Cost Management connector for Power BI – Ideal for small accounts with limited needs. Not recommended if reporting on more than $2M in total costs.
  • Cost Management exports in storage – Requires exporting data from Cost Management into a storage account. Doesn't require other deployments.
  • FinOps hubs – Requires deploying the FinOps hub solution.

If you need assistance in choosing the best approach for your needs, see Choosing a Power BI data source.

If using the Cost Management connector, see Create visuals and reports with the Cost Management connector in Power BI Desktop.

If using FinOps hubs, you can copy the URL from the deployment outputs in the Azure portal:

  1. Navigate to the resource group where FinOps hubs was deployed.
  2. Select Settings > Deployments in the menu.
  3. Select the hub deployment.
  4. Select Outputs in the menu.
  5. Copy the storageUrlForPowerBI value.
  6. Paste this URL into the Hub storage URL in Power BI.
  7. If using raw exports for any data, also use the following steps.
  8. If not using raw exports for any data, paste the hub storage URL into the Export storage URL in Power BI.

    Note

    Power BI requires both parameters to be set in order for the Power BI service to refresh datasets.

If using raw exports without FinOps hubs for any datasets (even if you're using hubs for cost data), you can obtain the Data Lake Storage URI from your storage account in the Azure portal:

  1. Navigate to the storage account in the Azure portal.
  2. Select Settings > Endpoints in the menu.
  3. Copy the Data Lake Storage > Data Lake Storage URL.
  4. Paste this URL into the Export storage URL in Power BI.
  5. If using FinOps hubs for any data, also follow the preceding steps.
  6. If not using FinOps hubs for any data, paste the export storage URL into the Hub storage URL in Power BI.

    Note

    Power BI requires both parameters to be set in order for the Power BI service to refresh datasets.

Step 2: Connect Power BI to storage

Decide whether want to connect to storage using a user or service principal account or using a storage account key (also called SAS token).

  • Using a user or service principal account
    1. Ensure you have the Storage Blob Data Reader role explicitly to the account to use. This permission isn't inherited even if you have "Owner" or "Contributor" permissions.
  • Using a SAS token
    1. Ensure you set the following permissions for the token:
      • Allowed services: Blob
      • Allowed resource types: Container and Object
      • Allowed permissions: Read and List
    2. Ensure you have also set a valid start and expiry date/time.

Step 3: Troubleshoot connection errors

  1. If you try to connect to your storage account and receive the Access to the resource is forbidden error, it's likely you're missing a few permissions. To ensure you have the correct permissions, refer back to Connect Power BI to storage.
  2. If you see an error about access being forbidden, review if the billing account that you're connecting to is correct. Power BI reports are provided with a sample billing account, and if you don't change that to your own ID, you can't connect.

Step 4: Troubleshoot missing months of data

  1. If the Power BI report doesn't include entire months of data, confirm the date parameters in the Power BI report by checking Transform data > Edit parameters in the ribbon. See Set up your first report for details.
    • Number of Months defines how many closed months (before the current month) get shown in reports. Even if data is exported, data outside this range isn't shown. If defined, this parameter overrides others.
    • RangeStart and **RangeEnd define an explicit date range of data to show in the reports. Anything before or after these dates isn't shown.
    • If RangeStart is empty, all historical data before RangeEnd is included.
    • If RangeEnd is empty, all new data after RangeStart is included.
    • If all date parameters are empty, all available data is included.

Related solutions: