Hi Yin, Sue,
Thanks for bringing this up. This behavior is expected and is related to the Azure AI Language Service model lifecycle, not an issue with your orchestration workflow itself.
The expiry date shown in Language Studio corresponds to the training configuration version used by your orchestration workflow (for example, 2022-09-01). Each training configuration has a defined lifecycle that includes a training expiration date and a deployment expiration date.
Once the training configuration expires, you will no longer be able to train or retrain models using that configuration version. However, any model that was already trained and deployed can continue running until the deployment expiration date.
When the deployment expiration date is reached, the deployed orchestration workflow will stop serving predictions unless you retrain the model using a newer supported configuration and redeploy it.
To avoid disruption, it’s recommended to monitor new training configuration versions, retrain your orchestration workflow with the latest supported version before expiration, and then redeploy it. This will refresh the lifecycle dates and keep your workflow operational.
In short: the workflow isn’t being removed automatically the expiry simply indicates when you’ll need to update and redeploy using a newer model configuration to continue using the service.
Hope this helps clarify things! Please let me know if there are any remaining questions or additional details, I can help with, I’ll be glad to provide further clarification or guidance.
Thankyou!