The server had an error while processing your request. Sorry about that!

Hello @Lunaar Vision ,

Welcome to Microsoft Q&A .Thank you for reaching out to us.

Based on the information provided,the error returned: “server_error” with a null code indicates that the request reached the service successfully but failed during backend processing before a detailed error could be generated.

This type of failure is typically associated with transient infrastructure conditions, such as:

• temporary backend instability
• internal timeouts during processing
• regional capacity or routing fluctuations

Since the same integration has been functioning consistently for several months, this strongly suggests that the issue is not related to request structure, authentication, or application logic but rather to a short-lived service-side condition.

Cloud-based AI services dynamically manage load, capacity, and infrastructure. Variations in demand, regional utilization, or temporary service events can occasionally result in intermittent server-side failures, even when no changes are made to the application.

Service health incidents or elevated error rates can also occur intermittently and get automatically mitigated.

These errors are commonly transient and recover automatically.

• Retry the request after a short interval
• Implement exponential backoff retry logic to handle temporary failures gracefully

Retrying failed requests with increasing delay intervals is a recommended best practice for handling transient API errors such as timeouts and server failures.

Please check if the following steps help if the issue still persists -

1. Checking service and resource health
   1. Review status via Azure Service Health overview
   2. This provides visibility into active incidents, planned maintenance, and service-impacting events affecting resources
2. Monitoring resource-level health
   1. Use Azure Resource Health overview
   2. This helps determine whether the issue is platform-related or specific to a resource
3. Validating behavior consistency
4. Retry with a minimal request (reduced payload or simpler input)
   1. Confirm whether failures are intermittent or consistent
5. Reviewing recent changes Please check for recent updates involving:
   • SDK or client library versions
   • API version changes
   • Endpoint configuration
   • Authentication settings
   • Request payload structure
   • Traffic volume or concurrency levels
6. Reviewing request diagnostics Please check and let us know if any HTTP status codes 500, 503, 504 are observed

At this stage, the most likely causes are a transient service-side condition, regional capacity constraints, or a deployment-specific issue.

The following references might be helpful , please check them out

• What is Azure Service Health? - Azure Service Health | Azure Docs
• Azure Resource Health overview - Azure Service Health | Microsoft Learn
• Foundry Models sold by Azure - Microsoft Foundry | Microsoft Learn

Please let us know if the response was helpful.

Thank you

Don't worry, this isn't an issue with your code or setup! A generic server_error with a null code typically means the backend hosting infrastructure itself hit a transient hiccup, a temporary timeout, or an unhandled processing exception while handling the model call.

Since it was working smoothly for you before, the best approach is to give it a few minutes and try executing the request again. If you're running this inside a production app, it’s a good reminder to make sure you have standard exponential backoff and retry logic implemented to automatically catch and smoothly handle these brief backend blips when they pop up.