This question already has answers here:
Execute task in background in WPF application
(3 answers)
Closed 2 years ago.
I am monitoring devices, using the example code below, and all works fine. The question is about how to handle each process at the same time. Let me explain.
When monitoring devices, we have an event that gets fired whenever a device disconnect. If for example we have to wait for each device to restart, or to reconnect to network, we use a thread.sleep() before starting the next commands. Only issue is if we have a large number of devices, each device will be done one at the time, potentially taking a long time to complete all devices. In this example, with a 10 seconds sleep, only 6 devices can be done every minutes..
How should I go to start 1 separate process for each device, and run it (almost) simultaneously? Or?
monitor.DeviceDisconnected += this.OnDeviceDisconnected;
private void OnDeviceDisconnected(object sender, DeviceDataEventArgs e)
{
......
Thread.Sleep(10000);
......
}
I would assume that the event is raised on the main thread since the UI freezes. As #Clemens mentions, you should avoid using Thread.Sleep since this will block the thread, and this is quite disrupting if done on the main thread.
To use await the method needs to be marked with async. Async methods need to either return a Task, Task<T> or void. It is generally recommended to return a Task since this lets the caller know when the method completes and if any exceptions occurred.
Since it is an event handler the caller would not care when the task completes, but we should take care to not lose any exceptions . So the method signature would look something like this
private async voidOnDeviceDisconnected(object sender, DeviceDataEventArgs e){
{
......
try{
await Task.Delay(TimeSpan.FromSeconds(10));
....
}
catch(Exception e){
// Handle exception
}
}
Related
I feel that the answer to this is due to me having an incorrect concept of how threads work, but here goes.
private void button1_Click(object sender, EventArgs e)
{
this.TestMethodAsync(); // No await, i.e. fire and forget
// ** Some code here to perform long running calculation (1) **
}
private async Task TestMethodAsync()
{
// Some synchronous stuff
await Task.Delay(1000);
// ** Some code here to perform long running calculation (2) **
}
First of all, I would not "fire and forget" an asynchronous method like this (I would use Task.Run) but I've come across code that does, and I'm trying to understand what the effect is.
In a WinForms application, which uses a WindowsFormsSynchronizationContext, my understanding of async and await tells me that when I click button1, the method will start synchronously on the UI thread. It will call TestMethodAsync and run synchronously until it reaches the await. It will then capture the context, start the Task.Delay task, and yield control to the caller. Since we are not awaiting this call, button1_Click will continue on the UI thread and start performing calculation (1).
At some point, Task.Delay(1000) will complete. A continuation will then run the remainder of the TestMethodAsync method using the captured context, which in this case means that the continuation will be run on the UI thread. This will now start performing calculation (2).
We now have two separate sections of code wanting to run on the same thread (the UI thread) at the same time. My investigations into this seem to suggest that the thread switches back and forth between the two sections of code in order to perform them both.
QUESTION:
I'm confused about exactly what is going on here. How is it possible to resume on a thread that is already running other code? What forces the thread to switch between the two sections of code that want to run? In general, what happens when you attempt to resume on a thread that is already running some other code?
(I suppose this isn't any different to how my click event runs on the UI thread in the first place, in as much as I know it runs on the UI thread, and I know the UI thread is also doing other stuff, but I've not really thought about it like this before.)
This is the secret that you do not understand: I give you the Windows Message Loop
int WINAPI WinMain(HINSTANCE hInstance, HINSTANCE hPrevInstance, LPSTR lpCmdLine, int nCmdShow)
{
MSG msg;
BOOL bRet;
while(TRUE)
{
bRet = GetMessage(&msg, NULL, 0, 0);
if (bRet <= 0) break;
TranslateMessage(&msg);
DispatchMessage(&msg);
}
return msg.wParam;
}
This is the actual "main" of your application; you just don't see it because it is hidden behind the scenes.
A simpler loop could not be imagined. It gets a message from the queue. If there are no more messages then the program must be done. If there was a message then it runs the standard message translations and dispatches the message, and then keeps on running.
How is it possible to resume on a thread that is already running other code?
It isn't. "Resuming on a thread that is running other code" is actually putting a message in the queue. That "other code" is being synchronously called by DispatchMessage. When it is done, it returns to the loop, the queue is polled, and the message indicates what code needs to be dispatched next. That then runs synchronously until it returns back to the loop.
What forces the thread to switch between the two sections of code that want to run?
Nothing. That doesn't happen.
In general, what happens when you attempt to resume on a thread that is already running some other code?
The message that describes what continuation needs to be run is queued up.
I suppose this isn't any different to how my click event runs on the UI thread in the first place, in as much as I know it runs on the UI thread, and I know the UI thread is also doing other stuff, but I've not really thought about it like this before.
Start thinking about it.
Click events are exactly the same. Your program is doing something; you click the mouse; the click hander does not interrupt the UI thread and start running new work on it. Rather, the message is queued up, and when your UI thread control returns to the message loop, the click is eventually processed; DispatchMessage causes Button1_OnClick to be invoked via some mechanism in Windows Forms. That's what WinForms is; a mechanism for translating Windows messages into calls to C# methods.
But you already knew that. You know that when an event-driven program does a long-running synchronous operation, that the UI freezes, but that click events are processed eventually. How did you think that happened? You must have understood at some level that they were being queued up for processing later, right?
Exercise: What does DoEvents do?
Exercise: Given what you now know: what could possibly go wrong if you call DoEvents in a loop to unblock your UI?
Exercise: How is await different from DoEvents in a GUI application?
How is it possible to resume on a thread that is already running other code?
It needs to be specifically designed to support it. There needs to be some framework in place that allows the thread to take in work and to then execute that work at some later point in time.
This is how your UI thread works. It has a queue, and whenever you schedule work to be done in the UI thread you add an item to the end of the queue. The UI thread then takes the first item from the queue, executes it, and then when it's done, goes on to the next item, and so on, until you end your application.
What forces the thread to switch between the two sections of code that want to run?
Nothing, because it doesn't do that. It runs one, then when it finishes, it runs the other.
In general, what happens when you attempt to resume on a thread that is already running some other code?
Either someone wrote some custom code to specifically do just that, in which case, it does whatever that code specifically told it to do, or else you can't.
I have a method that utilises SendKeys.Send, waits for a few seconds using System.Threading.Thread.Sleep and then runs another method to check the colour of a pixel to see if it has changed. The method then runs again as it is called recursively.
This method needs to be able to run thousands of times before stopping. The Winform's UI seems to stop responding when this is running.
I tried to implement a background worker to take the strain off of the UI. I moved the recursive method's code over to the Do_Work event and called it with RunWorkerAsync but it crashed, reporting the following:
An exception of type 'System.InvalidOperationException' occurred in System.Windows.Forms.dll but was not handled in user code
Additional information: SendKeys cannot run inside this application because the application is not handling Windows messages.
What's the best way of moving the code away from the UI? I'm not very familiar with background workers so I may be doing it wrong.
Sounds like a case for async. Try replacing Thread.Sleep() with Task.Delay().
async void Button_Click(object sender, RoutedEventArgs e)
{
await SendMyKeysAsync();
}
async Task SendMyKeysAsync()
{
while (thePixelIsStillRed)
{
SendKeys.Send("whatever");
await Task.Delay(TimeSpan.FromSeconds(1));
}
}
This approach leaves the UI thread free to continue pumping messages during the delay period, without spawning any additional threads.
Rather than a synchronous recursive method, you should write an asynchronous iterative method.
private async void Foo()
{
while(ShouldKeepLooping())
{
SendKeys.Send(keyToSend);
await Task.Delay(timespan.FromSeconds(2));
}
}
Making the method recursive adds nothing; making it iterative removes stack pressure. By making the method asynchronous, rather than synchronous, you don't block the UI thread.
Sorry for the lengthy post, I just want to illustrate my situation as best as possible. Read the items in bold and check the code if you want the quick gist of the issue.
I use ClickOnce to deploy a C# application, and have opted to have my application check for updates manually using the ApplicationDeployment Class rather than letting it do the update checking for me.
The program is a specialized network scanner that searches for network devices made by the company I work for. Once the main window is loaded, a prompt is displayed asking if the user would like to scan the network. If they say Yes, a scan begins which can take a minute or two to complete depending on their network settings; otherwise it just waits for the user to do some action.
One of the last things I do in Form_Load is create a new thread that checks for updates. This had all been working fine for several months through about 12 releases and has suddenly stopped working. I didn't change the update code at all, nor change the sequence of what happens when the app starts.
In staring at the code, I think I see why it is not working correctly and wanted to confirm if what I think is correct. If it is, it begs the question as to why it DID work before - but I'm not too concerned with that either.
Consider the following code:
frmMain.cs
private void Form1_Load(object sender, EventArgs e)
{
// set up ui, load settings etc
Thread t = new Thread(new ParameterizedThreadStart(StartUpdateThread));
t.Start(this);
}
private void StartUpdateThread(object param)
{
IWin32Window owner = param as IWin32Window;
frmAppUpdater.CheckForUpdate(owner);
}
frmAppUpdater.cs
public static void CheckForUpdate(IWin32Window owner)
{
if (ApplicationDeployment.IsNetworkDeployed) {
Console.WriteLine("Going to check for application updates.");
parentWindow = owner;
ApplicationDeployment ad = ApplicationDeployment.CurrentDeployment;
ad.CheckForUpdateCompleted += new CheckForUpdateCompletedEventHandler(ad_CheckForUpdateCompleted);
ad.CheckForUpdateProgressChanged += new DeploymentProgressChangedEventHandler(ad_CheckForUpdateProgressChanged);
ad.CheckForUpdateAsync();
// CAN/WILL THE THREAD CREATED IN FORM1_LOAD BE TERMINATED HERE???
}
}
When the CheckForUpdateAsync() callback completes, if no update is available the method call simply returns; if an update IS available, I use a loop to block until 2 things occur: The user has dismissed the "Would you like to scan prompt" AND no scan is currently running.
The loop looks like this, which takes place in ad_CheckForUpdateCompleted:
while (AppGlobals.ScanInProgress || AppGlobals.ScanPromptVisible) {
System.Threading.Thread.Sleep(5000);
}
I sleep for 5 seconds because I figured this was happening in a separate thread and it has seemed to work well for a while.
My main question about the above code is:
When ad.CheckForUpdateAsync(); is called from CheckForUpdate does the thread I created in Form1_Load terminate (or might it terminate)? I suspect it may because the subsequent Async call causes the method to return, and then start another thread?
The only reason I am confused is because this method WAS working for so long without hanging the application and now all of the sudden it hangs and my best effort at debugging revealed that it was that Sleep call blocking the app.
I'd be happy to post the full code for frmAppUpdater.cs if it would be helpful.
When ad.CheckForUpdateAsync(); is called from CheckForUpdate does
the thread I created in Form1_Load terminate (or might it terminate)?
If the CheckForUpdateAsync() call is asynchronous then yes, the thread will terminate, no it won't otherwise.
If you suspect the Sleep to have caused the application hang then these two variables AppGlobals.ScanInProgress and AppGlobals.ScanPromptVisible are probably always set to true! You should start looking at the code that is setting them to true and see what is going on there.
In order to avoid an application hang, you could introduce a variable to avoid sleeping indefinitely:
int nTrials = 0;
while ((AppGlobals.ScanInProgress || AppGlobals.ScanPromptVisible) && (nTrials < 5)) {
System.Threading.Thread.Sleep(5000);
nTrials++;
}
// Check the results and act accordingly
I personally do not like using Sleep for thread synchronization. .NET offers a bunch of classes that are perfect for thread synchronization, WaitHandle being one of them.
See this post at Asynchronous Delegates Vs Thread/ThreadPool?
your form load method seems to be doing synchronous work. you mention that you are using clickonce deployment. Has the binary location changed after the previous release or has permissions on this resource changed. Looks like the work (checkupdates) in the Thread is never finishing and is never handed back to the form.
as an immediate fix, I would change the Thread approach to Delegate - if you use delegate, then this becomes less of a customer issue (the form will respond to end user) but the underlying problem remains.
as the next step, i would go through http://msdn.microsoft.com/en-us/library/ms229001.aspx and do the troubleshoot
Whenever i am updating UI in windows form using delegate it gives me cross thread exception
why it is happening like this?
is there new thread started for each delegate call ?
void Port_DataReceived(object sender, SerialDataReceivedEventArgs e)
{
//this call delegate to display data
clsConnect(statusMsg);
}
protected void displayResponse(string resp)
{
//here cross thread exception occur if directly set to lblMsgResp.Text="Test";
if (lblMsgResp.InvokeRequired)
{
lblMsgResp.Invoke(new MethodInvoker(delegate { lblMsgResp.Text = resp; }));
}
}
The DataReceived event is always raised on a threadpool thread. You cannot update any UI control, you have to use Control.BeginInvoke(). There is no point testing InvokeRequired, it is always true.
A couple of things to keep in mind here:
Don't call Control.BeginInvoke for every single character or byte that you receive. That will bring the UI thread to its knees. Buffer the data you get from the serial port until you've got a complete response. Using SerialPort.ReadLine() usually works well, a lot of devices send strings that are terminated by a line feed (SerialPort.NewLine).
Shutting down your program can be difficult. You have to make sure to keep the form alive until the serial port stops sending. Getting an event after the form is closed will generate an ObjectDisposed exception. Use the FormClosing event to close the serial port and start a one second timer. Only really close the form when the timer expires.
Avoid using Control.Invoke instead of BeginInvoke. It can deadlock your program when you call SerialPort.Close().
Lots of ways to get in trouble. Consider using your own thread instead using DataReceived to avoid them.
Port_DataReceived is obviously an async event handler that is being raised by a thread on port monitoring component.
is there new thread started for each
delegate call ?
No, probably not. Your port monitoring component is running the poll on a background thread and the event is being raised from that thread, every time.
The point is that it is being called on a thread other than the UI, so you will need to use Control.Invoke and the patterns associated with it.
Consider this, (and read the post that may illuminate things for you)
void Port_DataReceived(object sender, SerialDataReceivedEventArgs e)
{
//this call delegate to display data
UpdateTheUI(statusMsg);
}
private void UpdateTheUI(string statusMsg)
{
if (lblMsgResp.InvokeRequired)
{
lblMsgResp.BeginInvoke(new MethodInvoker(UpdateTheUI,statusMsg));
}
else
{
clsConnect(statusMsg);
}
}
With all of that said, I would be remiss if I didn't point out that the indirection is troubling.
Cross Thread exception happens when some none UI thread changes UI elements. Since UI elements should be changed only in the UI thread this exception is thrown. To help you understand why this happen you will have to publish you code.
Cross thread exception happens when some none UI thread changes the UI elements. To fix this use the Invoke method on the control itself. As an extra you can check InvokeRequired on the control before calling the Invoke method
See msdn
[This appears to be a loooong question but I have tried to make it as clear as possible. Please have patience and help me...]
I have written a test class which supports an Async operation. That operation does nothing but reports 4 numbers:
class AsyncDemoUsingAsyncOperations
{
AsyncOperation asyncOp;
bool isBusy;
void NotifyStarted () {
isBusy = true;
Started (this, new EventArgs ());
}
void NotifyStopped () {
isBusy = false;
Stopped (this, new EventArgs ());
}
public void Start () {
if (isBusy)
throw new InvalidOperationException ("Already working you moron...");
asyncOp = AsyncOperationManager.CreateOperation (null);
ThreadPool.QueueUserWorkItem (new WaitCallback (StartOperation));
}
public event EventHandler Started = delegate { };
public event EventHandler Stopped = delegate { };
public event EventHandler<NewNumberEventArgs> NewNumber = delegate { };
private void StartOperation (object state) {
asyncOp.Post (args => NotifyStarted (), null);
for (int i = 1; i < 5; i++)
asyncOp.Post (args => NewNumber (this, args as NewNumberEventArgs), new NewNumberEventArgs (i));
asyncOp.Post (args => NotifyStopped (), null);
}
}
class NewNumberEventArgs: EventArgs
{
public int Num { get; private set; }
public NewNumberEventArgs (int num) {
Num = num;
}
}
Then I wrote 2 test programs; one as console app and another as windows form app. Windows form app works as expected when I call Start repeatedly:
But console app has hard time ensuring the order:
Since I am working on class library, I have to ensure that my library works correctly in all app models (Haven't tested in ASP.NET app yet). So I have following questions:
I have tested enough times and it appears to be working but is it OK to assume above code will always work in windows form app?
Whats the reason it (order) doesn't work correctly in console app? How can I fix it?
Not much experienced with ASP.NET. Will the order work in ASP.NET app?
[EDIT: Test stubs can be seen here if that helps.]
Unless I am missing something then given the code above I believe there is no way of guaranteeing the order of execution. I have never used the AsyncOperation and AsyncOperationManager class but I looked in reflector and as could be assumed AsyncOperation.Post uses the thread pool to execute the given code asynchronously.
This means that in the example you have provided 4 tasks will be queued to the thread pool synchronously in very quick succession. The thread pool will then dequeue the tasks in FIFO order (first in first out) but it's entirely possible for one of later threads to be scheduled before an earlier one or one of the later threads to complete before an earlier thread has completed its work.
Therefore given what you have there is no way to control the order in the way you desire. There are ways to do this, a good place to look is this article on MSDN.
http://msdn.microsoft.com/en-us/magazine/dd419664.aspx
I use a Queue you can then Enqueue stuff and Dequeue stuff in the correct order. This solved this problem for me.
The documentation for AsyncOperation.Post states:
Console applications do not synchronize the execution of Post calls. This can cause ProgressChanged events to be raised out of order. If you wish to have serialized execution of Post calls, implement and install a System.Threading.SynchronizationContext class.
I think this is the exact behavior you’re seeing. Basically, if the code that wants to subscribe to notifications from your asynchronous event wants to receive the updates in order, it must ensure that there is a synchronization context installed and that your AsyncOperationManager.CreateOperation() call is run inside of that context. If the code consuming the asynchronous events doesn’t care about receiving them in the correct order, it simply needs to avoid installing a synchronization context which will result in the “default” context being used (which just queues calls directly to the threadpool which may execute them in any order it wants to).
In the GUI version of your application, if you call your API from a UI thread, you will automatically have a synchronization context. This context is wired up to use the UI’s message queueing system which guarantees that posted messages are processed in order and serially (i.e., not concurrently).
In a Console application, unless if you manually install your own synchronization context, you will be using the default, non-synchronizing threadpool version. I am not exactly sure, but I don’t think that .net makes installing a serializing synchronization context very easy to do. I just use Nito.AsyncEx.AsyncContext from the Nito.AsyncEx nuget package to do that for me. Basically, if you call Nito.AsyncEx.AsyncContext.Run(MyMethod), it will capture the current thread and run an event loop with MyMethod as the first “handler” in that event loop. If MyMethod calls something that creates an AsyncOperation, that operation increments an “ongoing operations” counter and that loop will continue until the operation is completed via AsyncOperation.PostOperationCompleted or AsyncOperation.OperationCompleted. Just like the synchronization context provided by a UI thread, AsyncContext will queue posts from AsyncOperation.Post() in the order it receives them and run them serially in its event loop.
Here is an example of how to use AsyncContext with your demo asynchronous operation:
class Program
{
static void Main(string[] args)
{
Console.WriteLine("Starting SynchronizationContext");
Nito.AsyncEx.AsyncContext.Run(Run);
Console.WriteLine("SynchronizationContext finished");
}
// This method is run like it is a UI callback. I.e., it has a
// single-threaded event-loop-based synchronization context which
// processes asynchronous callbacks.
static Task Run()
{
var remainingTasks = new Queue<Action>();
Action startNextTask = () =>
{
if (remainingTasks.Any())
remainingTasks.Dequeue()();
};
foreach (var i in Enumerable.Range(0, 4))
{
remainingTasks.Enqueue(
() =>
{
var demoOperation = new AsyncDemoUsingAsyncOperations();
demoOperation.Started += (sender, e) => Console.WriteLine("Started");
demoOperation.NewNumber += (sender, e) => Console.WriteLine($"Received number {e.Num}");
demoOperation.Stopped += (sender, e) =>
{
// The AsyncDemoUsingAsyncOperation has a bug where it fails to call
// AsyncOperation.OperationCompleted(). Do that for it. If we don’t,
// the AsyncContext will never exit because there are outstanding unfinished
// asynchronous operations.
((AsyncOperation)typeof(AsyncDemoUsingAsyncOperations).GetField("asyncOp", BindingFlags.NonPublic|BindingFlags.Instance).GetValue(demoOperation)).OperationCompleted();
Console.WriteLine("Stopped");
// Start the next task.
startNextTask();
};
demoOperation.Start();
});
}
// Start the first one.
startNextTask();
// AsyncContext requires us to return a Task because that is its
// normal use case.
return Nito.AsyncEx.TaskConstants.Completed;
}
}
With output:
Starting SynchronizationContext
Started
Received number 1
Received number 2
Received number 3
Received number 4
Stopped
Started
Received number 1
Received number 2
Received number 3
Received number 4
Stopped
Started
Received number 1
Received number 2
Received number 3
Received number 4
Stopped
Started
Received number 1
Received number 2
Received number 3
Received number 4
Stopped
SynchronizationContext finished
Note that in my example code I work around a bug in AsyncDemoUsingAsyncOperations which you should probably fix: when your operation stops, you never call AsyncOperation.OperationCompleted or AsyncOperation.PostOperationCompleted. This causes AsyncContext.Run() to hang forever because it is waiting for the outstanding operations to complete. You should make sure that your asynchronous operations complete—even in error cases. Otherwise you might run into similar issues elsewhere.
Also, my demo code, to imitate the output you showed in the winforms and console example, waits for each operation to finish before starting the next one. That kind of defeats the point of asynchronous coding. You can probably tell that my code could be greatly simplified by starting all four tasks at once. Each individual task would receive its callbacks in the correct order, but they would all make progress concurrently.
Recommendation
Because of how AsyncOperation seems to work and how it is intended to be used, it is the responsibility of the caller of an asynchronous API that uses this pattern to decide if it wants to receive events in order or not. If you are going to use AsyncOperation, you should document that the asynchronous events will only be received in order by the caller if the caller has a synchronization context that enforces serialization and suggest that the caller call your API on either a UI thread or in something like AsyncContext.Run(). If you try to use synchronization primitives and whatnot inside of the delegate you call with AsyncOperation.Post(), you could end up putting threadpool threads in a sleeping state which is a bad thing when considering performance and is completely redundant/wasteful when the caller of your API has properly set up a synchronization context already. This also enables the caller to decide that, if it is fine with receiving things out of order, that it is willing to process events concurrently and out of order. That may even enable speedup depending on what you’re doing. Or you might even decide to put something like a sequence number in your NewNumberEventArgs in case the caller wants both concurrency and still needs to be able to assemble the events into order at some point.