Pipeline activity Refresh SQL Endpoint (preview) does not accept dynamic content
I am testing out the new Pipeline activity for Refresh SQL Endpoint, currently in preview.
Sure, I get it that this is in preview, but it should at least have basic minimum functionality. It does not.
The Workspace ID property does not even allow Dynamic Content like a Copy Data activity Destination tab.
The Lakehouse Endpoint ID will accept Dynamic Content, but after saving and re-opening the pipeline, that Dynamic Content is empty.
Please fix these basic issues.
6 Comments
- Another question: This task references the "SQL Endpoint Id" property. But behind that SQL Endpoint are ALL the warehouses, lakehouses, and mirrored Fabric SQL databases. The NAME of the property is not consistent with the selected OPTIONS. Either: NAME the property something like "Lakehouse or Warehouse Name" or Make it so that I can target the SQL Analytic Endpoint associated with a Workspace (and we all know there is one and only one of those per workspace, right?) and have it refresh ALL objects in the Workspace.
- Is the functionality for this task going downhill? I added a task to a new pipeline and set the properties by selecting from the dropdown options. I save and close the pipeline. It gets called from a parent pipeline and when that happens, this task fails. So I open it again and the SQL endpoint Id is now EMPTY. So I select one of the entries. (Keep in mind that I am NOT doing Dynamic Content on this task.) Save and close, then re-open. EMPTY again. WTF?!? At one point, forgetting that Dynamic Content was USELESS for this task, I had used it. Only to find that the settings were lost (see above). But AFTER realizing that, I switched to selected items from the lists, and THOSE got lost as well. I DELETED the task and started over. Looks like NOW it is holding the settings through a SAVE and CLOSE. So, bottom line, once you specify Dynamic Content on this, you have hosed it up. Delete the task and start over. Painful lesson to learn over hours of debugging. Microsoft: Please FIX THIS!!!
Next issue: I have this task inside a For Each loop and so runs several times during pipeline. Most of the times it runs just fine. About a third of the time it fails with the following error: Error calling the endpoint 'https://api.fabric.microsoft.com'. Response status code: 'ClientSideException HttpStatusCode : 408, HttpStatusString: RequestTimeout'. More details: Exception message: 'NA - Unknown [ClientSideException] A task was canceled.'. Request didn't reach the server from the client. This could happen because of an underlying issue such as network connectivity, a DNS failure, a server certificate validation or a timeout. Url endpoint request timed out. Please make sure the endpoint response is within 1 minute and retry. What is going on? To mitigate this, I put in a Refresh and specify 2 minutes, and most of those second tries go OK.
- I got the same issue with the activity Refresh SQL Endpoint. When I try to set dynamic content, it says "SQL endpoint is required.". This should be resolved asap.
This activity has some serious issues. I too have encountered all of the issues listed above, and even some rather strange UI bugs where the field label changes to "SQL endpoint Id ID" (perhaps triggered when you enter a dynamic value). Also strange bugs once you have attempted to use a dynamic value, where even changing it back to a static SQL endpoint selected from the drop-down can change itself once the pipeline is saved and reopened. As someone else implied, once you have tried to use a dynamic value for SQL Endpoint ID, you are best off nuking the pipeline and starting again.
This is GA now isn't it (even that doesn't seem clear), so why does it have so many outstanding bugs reported months ago? There is no way that this activity can be used in any serious workload needing a deployment pipeline.
I have been having this same issue with Fabric deleting my settings on the Refresh SQL Endpoint activity too. I have tried with both dynamic content and static strings to no avail. This issue (I think) is also causing me problems when I want to deploy my pipelines from a development workspace to production one. The deploy fails with an error message saying "'WorkspaceId' cannot be null" when there is clearly a workspaceId before I deploy the pipeline. This issue needs fixing because I am relying on a Wait activity at the moment set to 20 minutes so that I can trust that the SQL endpoint has been refreshed... if I could rely on the refresh SQL endpoint then it would cut the time from 20 minutes to 9 seconds.
