Azure Function App (Python v2) - blob_trigger function not showing after deployment

Question

I have a Function App in a Python v2 programming model, which has a blob_trigger that is supposed to do sth on new blobs being created in a given storage account and container. I am working in VSCode. The function app works well when running locally with Azurite, however when I deploy it to Azure, there is no "blobTrigger" function under the Function App.

This is a code of my function_app.py:

import azure.functions as func
from azure.storage.blob.aio import BlobServiceClient

import datetime
import logging
import json
import os
import sys
import asyncio

_KV_SECRET_INPUT_BLOB_CONNECTION_STRING = os.environ["KV_SECRET_INPUT_BLOB_CONNECTION_STRING"]
STORAGE_CONTAINER_NAME = os.environ["STORAGE_CONTAINER_NAME"]

async def read_blob_contents(connection_string: str, input_container_name: str, input_blob_name: str):
    blob_service_client = BlobServiceClient.from_connection_string(connection_string)
    container_client = blob_service_client.get_container_client(input_container_name)
    blob_client = container_client.get_blob_client(input_blob_name)

    try:
        blob_contents = await blob_client.download_blob(encoding = 'utf-8')
        return await blob_contents.readall()
    except Exception as e:
        logging.error(f"Error reading blob: {str(e)}")
        return None
    finally:
        await blob_service_client.close()


app = func.FunctionApp()
@app.function_name(name="blobTrigger")
@app.blob_trigger(arg_name="myblob", path=STORAGE_CONTAINER_NAME, connection="KV_SECRET_INPUT_BLOB_CONNECTION_STRING")

async def blob_trigger(myblob: func.InputStream, context: func.Context) -> None:

    func_invoc_and_event_meta = json.dumps({
        "function_name" : context.function_name,
        "function_invocation_id" : context.invocation_id,
        'blob_name': myblob.name,
        'blob_length': myblob.length,
        'blob_uri': myblob.uri,
    }, indent=4)

    logging.info('Blob trigger processed an event:')
    logging.info(func_invoc_and_event_meta)

    logging.info("Attempting to read the blob...")
    
    try:
        blob_url = myblob.uri
        logging.info(f"blob_url: '{blob_url}'")
               
        # Read the contents of the blob asynchronously
        input_container_name = myblob.name.split('/', maxsplit=1)[0]
        input_blob_name = myblob.name.split('/', maxsplit=1)[1]
        blob_contents = await read_blob_contents(connection_string=_KV_SECRET_INPUT_BLOB_CONNECTION_STRING, input_container_name=input_container_name, input_blob_name=input_blob_name)
        logging.error("blob reading completed")

    except:
        logging.error("ERROR during blob read : ")
        logging.error(str(sys.exc_info()))

I made sure that the Function App has appropriate appsettings i.e. KV_SECRET_INPUT_BLOB_CONNECTION_STRING and STORAGE_CONTAINER_NAME. The KV_SECRET_INPUT_BLOB_CONNECTION_STRING setting contains a connection string to the storage account that aims to store the data (new blobs), and it is a different storage account than the one that is linked to the function app through the AzureWebJobsStorage appsetting.

So what could be the problem?

On a related note, I also deployed another function app which is triggered by CRON, and for that one everything worked as expected.

Answer 1

After testing and double checking everything I managed to narrow down issues to the following:

1. requirements.txt did not contain all the python libs that were referenced in the import statements in the function_app.py code
2. there was an application settings of the Function App that was misspelled in the application code in the os.environ statement, and as such could not be found

Fixing those fixed the problem. When deployed, the function correctly appears under the function app.

Answer 2

Try to:

1. Set FUNCTIONS_EXTENSION_VERSION to ~4. Check this article to guide you.
2. You have to use General Purpose storage account.

Answer 3

• If you have the setting "AzureWebJobsStorage": "UseDevelopmentStorage=true" in local.settings.json, add the storage connection string as its value instead of "UseDevelopmentStorage=true".

• Add the setting "AzureWebJobsFeatureFlags": "EnableWorkerIndexing" in local.settings.json and in your Function App Environment variables.

---

I have created a Python V2 Azure function and deployed to Azure Function App with Consumption Plan.

local.settings.json:

{
  "IsEncrypted": false,
  "Values": {
    "FUNCTIONS_WORKER_RUNTIME": "python",
    "AzureWebJobsFeatureFlags": "EnableWorkerIndexing",
    "AzureWebJobsStorage": "<storage_connection_string>"
      }
}

• Deployment Status:

Portal: