Logic Apps Agent returns InternalServerError and retry policy cannot be saved

Question

Logic Apps Agent returns InternalServerError and retry policy cannot be saved

Ruoxuan Zhang (LTIMindtree Limited) 20 Microsoft External Staff

Issue 1 – Runtime failure

The workflow generally executes successfully.

However, during some executions, the Agent action fails with an InternalServerError.

This causes the workflow run to fail even though no changes were made to the workflow or its inputs.

The error appears to be intermittent and cannot be reproduced consistently. Issue 1 – Runtime failure
- The workflow generally executes successfully.
- However, during some executions, the Agent action fails with an InternalServerError.
- This causes the workflow run to fail even though no changes were made to the workflow or its inputs.
- The error appears to be intermittent and cannot be reproduced consistently.

Issue 2 – Retry policy configuration

As a mitigation, we attempted to configure a retry policy on the Agent action.

However, when saving the workflow after adding the retry policy, the Logic Apps designer itself returns an InternalServerError, preventing the retry policy from being saved.

As a result, we cannot implement the recommended retry mechanism to mitigate the intermittent runtime failures. Issue 2 – Retry policy configuration
- As a mitigation, we attempted to configure a retry policy on the Agent action.
- However, when saving the workflow after adding the retry policy, the Logic Apps designer itself returns an InternalServerError, preventing the retry policy from being saved.
- As a result, we cannot implement the recommended retry mechanism to mitigate the intermittent runtime failures.

Siddhesh Desai 7,890 Reputation points Microsoft External Staff Moderator

2026-07-21T23:35:48.3533333+00:00

Hi @Ruoxuan Zhang (LTIMindtree Limited)

Thank you for reaching out to Microsoft Q&A.

Based on the behavior described, there appear to be two separate issues affecting the Logic App Agent action.

For the runtime failure, the Agent action intermittently returns an InternalServerError even though the workflow configuration and inputs remain unchanged. Intermittent 5xx failures are typically indicative of a transient service-side or backend dependency issue rather than a workflow configuration problem. Since the same workflow succeeds in other executions without modification, this suggests that the failure is occurring within the Agent service or one of its downstream dependencies during execution. Microsoft documentation recommends using retry policies for transient failures when supported by the action. [Handle Wor...soft Learn | Learn.Microsoft.com]

For the retry policy configuration issue, the Logic Apps designer itself returns InternalServerError when attempting to save the Agent action after adding a retry policy. Similar save-time validation and designer-side InternalServerError issues have been identified previously in Logic Apps Agent-related scenarios and were acknowledged as product bugs by the Logic Apps engineering team.

Refer below points for mitigation and next steps:

Capture failed run details

Collect the Workflow Run ID, Correlation ID, Tracking ID, timestamp, and region from a failed execution. Compare successful and failed runs to determine whether the failure occurs consistently at the same Agent step.

Validate whether the Agent action supports retry policies

Not all Logic Apps actions support custom retry configuration. Microsoft documentation recommends first confirming that the specific trigger/action exposes retry settings.

Test in Code View

If the designer cannot save the retry policy, attempt to add the retryPolicy definition directly in the workflow JSON and validate whether the workflow can be saved and deployed.

If the save still fails, this further indicates a product-side validation issue rather than a workflow configuration problem.

Try an alternate portal session

Perform a hard refresh (Ctrl+Shift+R), use an InPrivate window, or test from another browser to rule out cached designer artifacts. Similar Logic Apps designer issues have previously been mitigated after portal-side fixes and refreshes.
Ruoxuan Zhang (LTIMindtree Limited) 20 Reputation points Microsoft External Staff

2026-07-24T19:02:39.58+00:00

I've tried saving the agent retry policy using both the Designer and the JSON code view, but neither approach works. If this is a known product-side validation issue, is there a planned fix? If so, is there an estimated timeline?

I'm including the workflow ID, relevant workflow details, and the error tracking ID below. Please use this information to help investigate the issue if possible. If any additional information is needed, I'm happy to provide it.

Workflow id: /subscriptions/fca2e8ee-1179-48b8-9532-428ed0873a2e/resourceGroups/Nexus_Package_Hosting/providers/Microsoft.Logic/workflows/Grafana_Newsletter_ADO_v2

Workflow start timestamp: 7/16/2026, 1:15:43 PM (UTC) The failed agent start timestamp: 7/16/2026, 1:24:19 PM (UTC)

Region: North Central US

Run id: 08584173991423793279687874481CU29

Error tracking id (for internal server error encountered during execution): 7ca88cc8-1c00-46ef-948e-861288201cd5 Error tracking id (for internal server error when saving the workflow with retry policy): 1bdb16cb-6601-49dc-ae3d-847e9f75cc98)

Answer accepted by question author

0 additional answers

Your answer

Ruoxuan Zhang (LTIMindtree Limited) 20 Reputation points Microsoft External Staff

2026-07-24T19:02:39.58+00:00

I've tried saving the agent retry policy using both the Designer and the JSON code view, but neither approach works. If this is a known product-side validation issue, is there a planned fix? If so, is there an estimated timeline?

I'm including the workflow ID, relevant workflow details, and the error tracking ID below. Please use this information to help investigate the issue if possible. If any additional information is needed, I'm happy to provide it.

Workflow id: /subscriptions/fca2e8ee-1179-48b8-9532-428ed0873a2e/resourceGroups/Nexus_Package_Hosting/providers/Microsoft.Logic/workflows/Grafana_Newsletter_ADO_v2

Workflow start timestamp: 7/16/2026, 1:15:43 PM (UTC) The failed agent start timestamp: 7/16/2026, 1:24:19 PM (UTC)

Region: North Central US

Run id: 08584173991423793279687874481CU29

Error tracking id (for internal server error encountered during execution): 7ca88cc8-1c00-46ef-948e-861288201cd5 Error tracking id (for internal server error when saving the workflow with retry policy): 1bdb16cb-6601-49dc-ae3d-847e9f75cc98)

Answer 1

hi Ruoxuan Zhang (LTIMindtree Limited) & thx for sharing urs issue here at Q&A portal,

look like two separate service-side failures. The intermittent InternalServerError from the Agent action is most likely a transient backend issue. A retry policy would normally be the right mitigation for a 5xx response, but only when that action supports configurable retries. Logic Apps documentation confirms that retry policies apply to supported actions and can also be added through code view.

The save operation returning another InternalServerError is the more important issue. That means the designer or workflow validation backend is rejecting the configuration before it can even be deployed. Try adding the retryPolicy directly in code view. If saving still fails, this isn’t something that can be fixed in the workflow logic.

Microsoft will need the failed run ID, workflow save correlation ID, exact UTC timestamps, region, Logic App type, and the Agent action definition. They should check both the Agent runtime telemetry and the Microsoft.Logic workflow validation logs.

rgds,

Alex

&

If my answer was helpful pls mark it and additional thx if u follow me at Q&A portal

and at my blog https://ctrlaltdel.blog/