队列处理后ASP.NET Core WebApi返回响应

Question

I have an endpoint which takes some entity as POST called Entry. These entries have a property called OrderNumber which is a number from 1-n determined based on entries that already happened today. So when there are 100 entries in the database for the current day, the new OrderNumber should be 101.

So what happens when you hit the endpoint twice from different devices (X and Y):

X. Sends an entry

X. Asks database for current entries -> response with 100

Y. Sends an entry

Y. Asks database for current entries -> response with 100

X. Record will be inserted with OrderNumber 101

Y. Record will be inserted with OrderNumber 101

My solution is to create a background task with a queue (as seen here: https://docs.microsoft.com/en-us/aspnet/core/fundamentals/host/hosted-services?view=aspnetcore-2.1#queued-background-tasks ) and create the entry there to ensure they are processed one after another.

Now the problem in my controller is, that I can't return the latest entry, with the proper OrderNumber during the request, because it happens whenever the queue has processed this entry.

So how does the client get the response back properly? Or is there a better/different way to handle these cases?

Answer 1

Custom attribute code here :

public class SingleThreadAttribute : ActionFilterAttribute
{
    private static Object _thisLock = new Object();

    public override void OnActionExecuting(ActionExecutingContext context)
    {
        Monitor.Enter(_thisLock);
        base.OnActionExecuting(context);
    }

    public override void OnActionExecuted(ActionExecutedContext context)
    {
        base.OnActionExecuted(context);
        Monitor.Exit(_thisLock);
    }
}

By adding [SingleThread] attribute on your controller's action endpoint, if multiple calls are made at the same time, they would be process one at the time waiting for the lock to be released by the previous call.

This could be a solution to your case.

Answer 2

I just found this stackoverflow answer which seem to resolve my issue:

How to protect resources that may be used in a multi-threaded or async environment?