503 server is unavailable even after successful deployment

Question

We have deployed a LLM application in web app service. the deployment was successful, but when we tested the url. it is throwing "503 server is unavailable".
The libraries used in the app is as follows:

fastapi==0.115.6
uvicorn==0.27.0
azure-search-documents==11.5.2
azure-core==1.32.0
openai==1.58.1
msal==1.31.1
websockets==14.1
azure-core==1.32.0
typing_extensions==4.12.2
aiohttp==3.11.11
transformers==4.47.1
bert-score==0.3.13
nltk==3.9.1
torch==2.5.1
numpy==2.2.1
python-dotenv==1.0.1
scikit-learn==1.6.0
streamlit==1.41.1
pandas==2.2.3
tqdm==4.67.1
rouge_score==0.1.2
scipy==1.14.1
spacy==3.8.3
selfcheckgpt==0.1.7
sentencepiece==0.2.0
https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.8.0/en_core_web_sm-3.8.0-py3-none-any.whl
accelerator_detect-0.1.1-py3-none-any.whl
azure-data-tables==12.6.0

We have custom whl file included as mentioned below, but this file is not creating any issue as we have tested the app with and without this file and getting the same error

accelerator_detect-0.1.1-py3-none-any.whl

In the platform logs, showing error as follows:

Please help us resolve the issue

Answer 1

@Raghunatha N S Thanks for posting your question in Microsoft Q&A, apologize for any inconvenience caused on this.

From the above shared logs, we understand that App service container not able to start within the specified time.

• Could you please add this WEBSITES_CONTAINER_START_TIME_LIMIT and value to 1800 to increase the app service container start time.
• Try disabling the HealthCheck on the app service and check,
• Review the app service logs by enabling the diagnostic logging to see if there are any application failures.
• Try troubleshooting the issue using the built-in detectors under the diagnose and solve problems of your app service.

Do let me know if you are still the facing the issue.