I have the requirement to execute an Azure Function (let’s call it OperationalFunction) only once in parallel. What I mean is the following:

I have two “entry point” functions:

I have a BlobTrigger Function (BlobTriggeredStartFunction)
I have a HttpTrigger Function (HttpTriggeredStartFunction)
Both of those functions call an orchestrator function (OrchestratorFunction) like this:

string instanceId = await starter.StartNewAsync(nameof(OrchestrationFunction),null,data);

This orchestration function has the following code:

[FunctionName("OrchestrationFunction")]
public async Task OrchestrationFunction(
[OrchestrationTrigger] IDurableOrchestrationContext context){
   var input = context.GetInput<string>();
   await context.CallActivityAsync(nameof(OperationalFunction),input);
}

I want that the *OperationalFunction *is executed ONLY ONCE in parallel, so that if I receive multiple triggers (n blob/http-triggered events), the *OperationalFunction *at a certain time, is executing only one instance, when such instance finishes, then the new event in the queue (FIFO) is executed.

Which is the best approach to that?

I tried to edit the host.json file by adding:

    "blobs": {
      "maxDegreeOfParallelism": 1,
      "poisonBlobThreshold": 1
    }

But it is not exactly what I am looking for, just a part of the problem is solved.

I also tried to add a loop do…while for getting the status of the orchestrator:

                do {
                    status = await starter.GetStatusAsync(instanceId);
                    await Task.Delay(2000);
                }
                while (
                    status.RuntimeStatus == OrchestrationRuntimeStatus.Running || 
                    status.RuntimeStatus == OrchestrationRuntimeStatus.Pending);

But I don’t think this is the most efficient solution.