Power Bi api throtteling becoming more aggresive, http error 429 (Too Many Requests)

[Remko1981NL]

We make extensive use of the Power Bi API to manage our reports and users. To that end once a week we extract a lot of meta data. Last month we ran into throtteling isseus. Therefore i added a sleep timer of 1 hour after a certain amount of requests. Unfortunatly this week the throtteling has become more severe. After only 200 api requests we get the http error 429 (Too Many Requests).

Does anyone else encounter this problem? Am i doing something wrong and therefore i get throttled more? Any ideas?

We use the api for the following data:

- get all report meta data

- get all workspaces

- get all reportusers

- get all Datasets

- get all Datasources

- Download all centrally produced reports

[Anonymous · Accepted solution]

Hi @Remko1981NL ,

Here are some suggestions that may help you alleviate this problem:

1. Try to optimize API calls. Ensure that the application only makes necessary API calls. Sometimes, applications make redundant requests that can be optimized. For example, if you frequently retrieve metadata that changes infrequently, consider caching this information and updating it at longer intervals.

2. Try Retry mechanism. For detailed information, please refer to the document: Azure service retry guidance - Best practices for cloud applications | Microsoft Learn.

3. Try leveraging the Power BI Administrator REST API for batch operations. For operations such as metadata extraction, consider using the admin REST API designed for batch operations.

4. Please check the following links similar to your question, I hope it will be helpful to you: Solved: Power BI Admin Too Many Requests - Microsoft Fabric Community.

If your Current Period does not refer to this, please clarify in a follow-up reply.

Best Regards,

Clara Gong

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

[Remko1981NL]

Thank you Clara for your response!

following the 4th link i found out that every API call has documentation on the maximum number of API calls per period under the limitations section. I should have better read the documentation.

Regarding your other suggestions:

1. Try to optimize API calls. Ensure that the application only makes necessary API calls. Sometimes, applications make redundant requests that can be optimized. For example, if you frequently retrieve metadata that changes infrequently, consider caching this information and updating it at longer intervals.

- I save all data in SQL server and only collect the information once a week.

2. Try Retry mechanism. For detailed information, please refer to the document: Azure service retry guidance - Best practices for cloud applications | Microsoft Learn.

- I believe this is not applicable if you use the Invoke-PowerBIRestMethod commandlet for Powershell. An improvement was suggested some time ago.

3. Try leveraging the Power BI Administrator REST API for batch operations. For operations such as metadata extraction, consider using the admin REST API designed for batch operations.

- Unfortunatly there are no batch options for the following:

Admin - Reports GetReportUsersAsAdmin - REST API (Power BI Power BI REST APIs) | Microsoft Learn

Admin - Datasets GetDatasourcesAsAdmin - REST API (Power BI Power BI REST APIs) | Microsoft Learn (needed because we have a live connection. Else i do not know what the source is)