REST API GET method Copy step ADF

Oscar Vicente Remon 0 Reputation points
2026-11-04T11:23:30.9433333+00:00

Hello since 29/10 we are experiencing the following error below in ADF copy steps with rest api and get method , nothing has been changed in our configuration and all set up is correct according to the documentation, including the pagination rules.

Failure happened on 'Source' side. ErrorCode=HttpInvalidUrlAddress,'Type=Microsoft.DataTransfer.Common.Shared.HybridDeliveryException,Message=Http Url Address 'https://wabi-north-europe-h-primary-redirect.analysis.windows.net' is invalid.,Source=Microsoft.DataTransfer.ClientLibrary,'

We would like to know why the system is not able to determine the pagination links and send us to this cluster since that specific day as nothing has been changed in our config , all set ups are working.

Is there a foreseen stable solution for this problem.

Azure Data Factory
Azure Data Factory

An Azure service for ingesting, preparing, and transforming data at scale.

0 comments No comments

3 answers

Sort by: Most helpful
  1. Hemant Bharadwaj 180 Reputation points Microsoft Employee
    2026-11-04T15:28:08.6533333+00:00

    Likely cause/summary: your API now returns a pagination value that ADF interprets as an invalid next URL (host-only redirect), so the Copy activity fails while composing the next page request. This usually impacts only pipelines that use pagination rules.

    1. In the REST source, avoid using AbsoluteUrl if the response field can return a partial/redirect host value. Use QueryParameters.* or Headers.* pagination rules based on the API's continuation token instead.
    2. Add an explicit end condition to stop paging safely (for example EndCondition:$.data = Empty, or the API-specific completion flag) and set MaxRequestNumber as a guardrail.
    3. Re-test the same request outside ADF (for example with curl/Postman) and verify the exact next-page value returned since 29/10.
    4. If the API really returns a full next-link, keep AbsoluteUrl but ensure the JSONPath points to the full URL field (including path/query), not a redirect host field.

    About your follow-up: pipelines that do not paginate typically should not be affected by this specific HttpInvalidUrlAddress issue, because they don't evaluate pagination rules.

    Reference docs:

    If you find this answer helpful, please upvote it. If this resolves your issue, please mark it as the accepted answer so that it can help others in the community.

    Was this answer helpful?

    0 comments No comments

  2. Oscar Vicente Remon 0 Reputation points
    2026-11-04T15:19:16.65+00:00

    can this affect other pipelines which use the same service but don't require pagination?

    Was this answer helpful?

    0 comments No comments

  3. AI answer

    2026-11-04T11:24:35.4133333+00:00

    Create a separate REST linked service whose base URL uses the regional wabi-north-europe-h-primary-redirect.analysis.windows.net host returned by the API, then run pagination through that connection so the initial and continuation requests remain on the same host.

    Apply the workaround

    A matching report began on October 29 with the same HttpInvalidUrlAddress failure when ADF followed a Power BI continuation URL to a regional wabi-...analysis.windows.net host. The reported workaround was to replace the generic Power BI base URL with the regional host used by the continuation URL ^1^.

    1. Run the initial request outside ADF and inspect its complete continuationUri. Record:
      • Scheme: https
      • Host: in your case, apparently wabi-north-europe-h-primary-redirect.analysis.windows.net
      • Common path prefix, such as the portion before the specific admin operation
    2. In ADF Studio → Manage → Linked services, clone the existing REST linked service.
    3. Change Base URL from the generic endpoint to the regional endpoint, preserving the common API path. For example, if the returned continuation URI is under /v1.0/myorg/admin/, use:
         https://wabi-north-europe-h-primary-redirect.analysis.windows.net/v1.0/myorg/admin/
      
      Use the exact host and path returned for your tenant; do not copy another region or cluster suffix.
    4. Keep the existing authentication configuration. Update the dataset’s relativeUrl so it contains only a path beneath that base URL. ADF combines the linked-service URL and dataset relative URL ^2^.
    5. Point the Copy activity to the cloned linked service and retain the existing pagination rule, for example:
         {
           "source": {
             "type": "RestSource",
             "requestMethod": "GET",
             "paginationRules": {
               "AbsoluteUrl": "$.continuationUri"
             }
           }
         }
      
    6. Run a debug execution and confirm that:
      • The first request succeeds.
      • The second-page URL has the same scheme, host, and port as the new base URL.
      • Multiple pages are copied.

    ADF officially supports obtaining the next absolute or relative URL from the current response. Pagination stops on HTTP 204 or when the configured JSONPath returns null ^2^.

    Stability and support status

    The matching public report says Microsoft Support attributed the behavior to a REST connector security update enforcing URL boundaries: continuation requests must remain within the base connection’s scheme, host, port, and path. That statement is reported publicly but is not currently documented as an official ADF connector rule ^1^.

    There is also no corresponding active entry on the public ADF known-issues page ^3^. Consequently:

    • The regional-base-URL approach is the best-matching operational workaround.
    • A rollback or permanent product fix is not publicly committed.
    • Because a regional cluster hostname could potentially change, it should not yet be treated as a guaranteed permanent endpoint unless Microsoft Support confirms that for your tenant.

    If the regional hostname later changes or the workaround fails, open an Azure support request under Data Factory → Copy Activity → Errors or unexpected results. Include the failed activity run ID, region, first response’s redacted continuationUri, old and new linked-service base URLs, and the first failure timestamp. Ask specifically whether the regional endpoint is supported as a durable base URL and whether the URL-boundary behavior will be formally documented.


    References

    1. FUAM Load_Activities_E2E Http Url Address 'https://wabi-north-europe-c ...
    2. Copy and transform data from and to a REST endpoint by using Azure Data Factory
    3. Azure Data Factory known issues - Azure Data Factory
    AI-generated content may be incorrect. Read our transparency notes for more information.

    Was this answer helpful?

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.