I have a WPF application which used to shutdown via Environment.Exit. Since this caused problems with my automated UI tests, I changed the exiting application to Application.Current.ShutDown.
This works fine, except there is a thread which is waiting to be pulsed (Monitor.Wait), which keeps the process running since it never gets pulsed anymore.
I thought that would be a no brainer to fix, but from the point where my application exits (the Application.Current.ShutDown) it's rather hard to retrieve a reference to the object which holds the waiting thread (in order to force pulse it, so that it can exit).
I tried to google an appropriate answer, but not much luck yet. Is there an "easy way" out of this? Or should I start refactoring already? :)
Some snippets:
Thread is created like this
workerThread = new Thread(Worker) { Name = logName, IsBackground = true};
In the Worker method, Monitor.Wait is called
while ((action = GetNextTask(out task)) == ProductPreparationAction.None)
{
Monitor.Wait(preparationTasks);
}
Nevermind my comment. Start refactoring :).
First of all, there should be a way for the while loop to end when before the app stops. Perhaps you can use and propagate a CancellationToken all the way down to the Worker method.
If you want to keep your loose coupling, you should be able to pulse by creating an event in the class that calls Application.Current.ShutDown and by subscribing to it in the class where the Worker method is (and call Pulse in the event handler).
If you do this, then you can store the CancellationToken in this class and flag it when the event is received.
That event should be raised before calling Application.Current.ShutDown.
Related
My program has 2 threads running, thread 1 does something to control a label in a form running on thread 2. So I have to use a delegate and invoke a function in form 1 class to access the label. My code is below and it works perfectly. However, I'm wondering if there is a shorter, better way to do this?
delegate void Change_Status_Call_Back(string status_changed);
public void change_status(string status_changed)
{
if (this.label_status.InvokeRequired)
{
Change_Status_Call_Back obj = new Change_Status_Call_Back(change_status);
this.Invoke(obj, new object[] { status_changed });
}
else
{
this.label_status.Text = status_changed;
}
}
This question is "primarily opinion based". Still, you've touched a pet peeve of mine, so…
You should skip the InvokeRequired check altogether:
public void change_status(string status_changed)
{
this.Invoke((MethodInvoker)(() => this.label_status.Text = status_changed));
}
The framework has to effectively check InvokeRequired anyway, because it's required to support invoking on the UI thread without deadlocking. So the check in your code is redundant. The overhead of always wrapping the method body in a delegate invocation is inconsequential in UI code like this, especially since if you're writing this code, it's probably the case that the method's not going to be called exception when InvokeRequired would be true anyway (i.e. the "fast path" is never taken anyway).
Even better is to use a more modern mechanism for dealing with cross-thread access, such as async/await or the Progress<T> class. Then you never have to write an explicit call to Invoke() at all.
Some time ago, I ranted in more depth here: MSDN’s canonical technique for using Control.Invoke is lame
I would do this:
public void change_status(string status_changed)
{
this.label_status.InvokeSafely(c => c.Text = status_changed);
}
You need this extension method:
public static void InvokeSafely(this Control control, Action<Control> action)
{
if (control.InvokeRequired)
{
control.Invoke((Action)(() => action?.Invoke(control)));
}
else
{
action?.Invoke(control);
}
}
After looking around, I came up with this:
// UPDATE DISPLAY items (using Invoke in case running on BW thread).
IAsyncResult h = BeginInvoke((MethodInvoker)delegate
{
FooButton.Text = temp1;
BarUpdown.Value = temp2;
}
);
EndInvoke(h); // Wait for invoke to complete.
h.AsyncWaitHandle.Close(); // Explicitly close the wait handle.
// (Keeps handle count from growing until GC.)
Details:
I removed if (InvokeRequired) completely. (Discovered from Peter Duniho's answer here.) Invoke() works just fine on the UI thread. In code that runs only on the UI thread, UI actions need no special treatment. In code that runs only on a non-UI thread, wrap all UI actions in an Invoke(). In code that can run on the UI thread -or- a non-UI thread, likewise wrap all UI actions in an Invoke(). Always using Invoke() adds some overhead when running on the UI thread, but: not much overhead (I hope); the actions run less often on the UI thread anyway; and by always using Invoke, you don't have to code the UI actions twice. I'm sold.
I replaced Invoke() with BeginInvoke() .. EndInvoke() .. AsyncWaitHandle.Close(). (Found elsewhere.) Invoke() probably just does BeginInvoke() .. EndInvoke(), so that much is just inline expansion (slightly more object code; slightly faster execution). Adding AsyncWaitHandle.Close() addresses something else: When running on a non-UI thread, Invoke() leaves hundreds of handles that linger until garbage collection. (It's scary to watch Handles count grow in Task Manager.) Using BeginInvoke() .. EndInvoke() leaves lingering handles just the same. (Surprise: Using only BeginInvoke() does not leave the handles; it looks like EndInvoke() is the culprit.) Using AsyncWaitHandle.Close() to explicitly kill the dead handles eliminates the [cosmetic] problem of lingering handles. When running on the UI thread, BeginInvoke() .. EndInvoke() (like Invoke()) leaves no handles, so AsyncWaitHandle.Close() is unnecessary, but I assume it is also not costly.
An IsDisposed test seems safe against race conditions, but I think it is not necessary. I'm worried that BackgroundWorker can Invoke() the operation; while it is pending, a click can trigger a callback on the UI thread that can Close() the form, and then the message loop executes this operation. (Not sure this can happen.)
Problem: (I will update here when something works.) I changed all my UI updates from running on a UI timer kludge to using Invoke() (as above), and now closing the form fails on a race condition about 20% of the time. If a user click stops my background worker, clicking on close after that works fine. BUT, if the user clicks directly on close, that triggers a callback on the UI thread which Close()s the form; that triggers another that flags the background worker to stop; the background worker continues, and it crashes at EndInvoke() saying "Cannot access a disposed object. Object name: 'MainWin'. at System.Windows.Forms.Control.MarshaledInvoke(Control caller, Delegate method, Object[] args, Boolean synchronous) ...". Adding if (!this.IsDisposed) {} around EndInvoke() .. AsyncWaitHandle.Close() doesn't fix it.
Option: Go back to using a forms timer: Make the BW write its changes into a dozen global "mailbox" variables. Have the timer do FooButton.Text = nextFooButtonText;, etc. Most such assignments will do almost nothing because setting a form field only updates the display if the value actually changes. (For clarity and to reduce copying objects, initialize the mailbox variables to null, and have the timer do if (nextFooButtonText != null) { FooButton.Text = nextFooButtonText; nextFooButtonText = null; }, etc.) The timer puts a new event on the UI message loop every so many milliseconds, which is more silly grinding than the Invoke()s. Updating the display on a timer callback delays each update by [up to] the timer interval. (Yuck.)
WORKING Option: Use only BeginInvoke(). Why make BW wait for each Invoke to finish? 1) temp1 and temp2 seem passed as references - if they get changed after BeginInvoke(), the new value wins. (But that's not so bad.) 2) temp1 and temp2 can go out of scope. (But aren't they safe against being released until the last reference goes away?) 3) Waiting ensures that BW only has one invoked action pending at a time - if the UI thread blocks for a while, BW can't bury it in events. (But my UI thread can't block, at least not at times when my BW is running.)
Option: Put try .. catch around the EndInvoke(). (Yuck.)
I have seen several other tricks suggested:
•Have Close cancel itself, initiate a timer, and then return so that any lingering Invoke()s finish on the UI thread; shortly after that the timer callback does a real Close (found here; from here).
•Kill the background worker thread.
•Alter Program.cs to shut down differently.
My application needs to review all new application Event Log entries as they come in.
private void eventLog_Application_EntryWritten(object sender, EntryWrittenEventArgs e)
{
// Process e.Entry
}
What I would like to know is what happens if another Entry is written to the EventLog while a previous Entry is being handled?
The documentation for EventLog.EntryWritten Event provides an example of handling an entry written event which uses threading (which is why I am asking the question).
In this example they use System.Threading and call the WaitOne() and Set() methods on the AutoResetEvent class, however I'm not sure precisely what this code is intended to achieve.
The documentation states that - WaitOne() "blocks the current thread until the current WaitHandle receives a signal", and that Set() "sets the state of the event to signaled, allowing one or more waiting threads to proceed". I'm not sure what the threading portion of this example is intended to demonstrate, and how this relates to how (or if) it needs to be applied in practice.
It appears that WaitOne() blocks the thread immediately after the entry has been written, until it has been handled, where it is then set to signalled (using Set()), before allowing the thread to proceed. Is this the one and only thread for the application?
Most importantly, when my application is not responsible for writing the the events which need to be read from the EventLog, how should this principle be applied? (If, indeed, it needs to be applied.)
What does happen if a new Entry is written while the application is inside the handler?
Nothing dramatic happens, it is serialized by the framework. The underlying winapi function that triggers the EventWritten event is NotifyChangeEventLog(). The .NET Framework uses the threadpool to watch for the event to get signaled with ThreadPool.RegisterWaitForSingleObject(). You can see it being used here.
Which is your cue to why the MSDN sample uses ARE (AutoResetEvent). The event handler runs on that threadpool thread, exactly when that happens is unpredictable. The sample uses a console mode app, without that ARE it would immediately terminate. With the ARE, it displays one notification and quits. Not actually that useful of course, I would personally just have used Console.ReadLine() in the sample so it just keeps running and continues to display info until you press the Enter key.
You don't need this if you use a service or a GUI app, something that's going to run for a long time until the user explicitly closes it. Note the EventLog.SynchronizingObject property, makes it easy to not have to deal with the threadpool thread in a Winforms app.
The example is not really helping to explain the way the AutoResetEvent works in a multi-threaded scenario, so I'll try to explain how I understand it to work.
The AutoResetEvent signal static variable, is instantiated as a new AutoResetEvent with its signaled state set to false, or "non-signaled", meaning that calling signal.WaitOne() will cause the thread that called WaitOne to wait at that point, until the signal variable is "set" by calling the signal.Set() method.
I found an explanation of AutoResetEvent that describes it very well in understandable real-world terms, which also included this excellent example below.
http://www.albahari.com/threading/part2.aspx#_AutoResetEvent
AutoResetEvent
An AutoResetEvent is like a ticket turnstile: inserting a ticket lets
exactly one person through. The “auto” in the class’s name refers to
the fact that an open turnstile automatically closes or “resets” after
someone steps through. A thread waits, or blocks, at the turnstile by
calling WaitOne (wait at this “one” turnstile until it opens), and a
ticket is inserted by calling the Set method. If a number of threads
call WaitOne, a queue builds up behind the turnstile. (As with locks,
the fairness of the queue can sometimes be violated due to nuances in
the operating system). A ticket can come from any thread; in other
words, any (unblocked) thread with access to the AutoResetEvent object
can call Set on it to release one blocked thread.
class BasicWaitHandle
{
static EventWaitHandle _waitHandle = new AutoResetEvent (false);
static void Main()
{
new Thread (Waiter).Start();
Thread.Sleep (1000); // Pause for a second...
_waitHandle.Set(); // Wake up the Waiter.
}
static void Waiter()
{
Console.WriteLine ("Waiting...");
_waitHandle.WaitOne(); // Wait for notification
Console.WriteLine ("Notified");
}
}
According to https://msdn.microsoft.com/en-us/library/0680sfkd.aspx the eventlog components are not thread-safe and that code is there to prevent unexpected behaviour in simultaneous interactions.
If multiple threads are executing these lines simultaneously, if is possible for one thread to change the EventLog.Source Property of the event log, and for another thread to write a message, after that property had been changed.
I have a weird issue:
In my C# app, I am creating another thread, like so:
Thread printThread = new Thread(printWorker);
printThread.Name = "Logger MainThread";
printThread.IsBackground = true;
printThread.Start();
When my main thread finishes, this new thread just keeps on working, although it's marked as Background.
What could be the causes for this?
This object is holding a Mutex object, not sure this may be the reason...
Any ideas anyone?
Here's the code from the printWorker method:
while (loggerIsActive)
{
LogMessage log = LoggerQueue.Dequeue();
if (log.message != null)
{
syncLogObj.WaitOne();
lock (writerobj)
{
StreamWriter sw;
if (!File.Exists(fName))
{
sw = File.CreateText(fName);
}
else
{
sw = new StreamWriter(fName, true);
}
using (sw)
{
if (log.message != "")
{
if (log.message.EndsWith("\r\n"))
{
log.message =
log.message.Substring(0, log.message.Length - 2);
}
sw.WriteLine(string.Format("[{0}][{3}][{1}] | {2}",
log.msgTime,
log.level.ToString(),
log.message,
log.sender.ToString()));
}
sw.Flush();
sw.Close();
}
}
syncLogObj.ReleaseMutex();
}
Thread.Sleep(5);
}
Try this:
Start the app through VS and exit normally. The VS should stay in Debug mode as you described. Click on Pause button (Break all) and then go to Debug->Windows->Threads. Do you see your "Logger MainThread" in the list?
If so, double-click it, it should lead you to the code line that the thread is currently executing. Step-debug from there and see why is it not terminating.
If you don't see it try looking at other threads that have not terminated and try to find the problem.
Otherwise, with those kind of problems it's always useful to monitor the program state via System.Diagnostics.Debug.Print statements (you can see them printing in the VS output window).
kill it.
Not pretty. But this isn't TV. Read on:
1) Not sure you use are using it but it appears you should be locking loggerqueue before you queue(main pgm) or dequeue(thread).
2) No need to lock writerobj with just this setting. But really you should so you can safely kill the thread not during a write:
main thread:
do everything
before close:
-lock writerobj
-printthread.abort
worker thread:
add try catch to handle threadabort exception and just quit
If you're properly doing this, you shouldn't have to use Waits and mutexes. If you are using wait properly anyway you won't need the sleep.
General advice for this application: why not log on main thread? if your logging is that busy, log results will be pretty useless.
But there are rare cases where that might be wrong. Entonces......
General advice to have threads play nice for this problem:
Main program
encapsulate logging (notably, quit flag, queue, and worker thread ref) in an object
'global snobs?' Logging is a rare excuse to use singleton patter.
start worker thread in logger object via method
main thread always calls a single method on logger object to log error
That method locks the queue and adds to it.
Use Monitor/Pulse/Wait, no sleep; full examples abound; it is worth learning
because only this thread is hitting the file anyway, unless you have multiple processes, you don't need waitone/releasemutex.
That logging method monitor.pulses an object
That frees the worker thread's monitor.wait (which is what idles the CPU instead of sleep)
lock the queue, only inside the lock dequeue the object to local ref; nothing else.
Do your normal logging code and 'exit check' loop. Add
Your logic code could leave message unwritten if queue is full on quit:
change to exit check so you can do it without an extra lock of queue:
move declaration of queued object refernce above while; set it to nothing
change logic in while to 'loggerisactive or log != null'
when your main thread finishes, in your exit code:
set the quit flag
pulse the object you're using to wait incase it's not processing the queue
Thread will fall thru.
You have a lot of stuff going on that you're obviously not showing...
Exmaple: you have syncLogObj.WaitOne();, but we don't see where syncLogObj is being declared, or used elsewhere in your program.
Plus, you don't need it... get rid of the syncLogObj thing altogether (including the "ReleaseMutex" garbage)... you already have a lock (blah) { }, and that's all you need (from what code you have displayed).
It's likely that the main thread is NOT ending, likely because of this or some other object that is keeping it open.
So, simple instructions
Get rid of syncLogObj (because you already have the "lock")
Make sure you set loggerIsActive = false somewhere.
Edit: Even more details!
From what I see - you don't need the lock (writerobj) at all, because (I'm quite sure), you only seem to have one thread that is writing to the log.
The "lock" is only there if you have two or more threads that running that code (basically).
If printworker does not finish before your main thread is done, then main will die and your printworker thread will be killed by the OS. If you want main to wait for the thread you created, then you should call printThread.Join() in main. That will get main to wait on your thread.
When main finishes your program dies and your printThread will be destroyed by the OS, It will not keep running.
From here
Background threads are identical to
foreground threads with one exception:
a background thread does not keep the
managed execution environment running.
Once all foreground threads have been
stopped in a managed process (where
the .exe file is a managed assembly),
the system stops all background
threads and shuts down.
Tony the Tiger has the right idea but additional code needs to be added to kill the thread before the application closes.
printThread.Join(1000);
if(printThread!=null && printThread.IsAlive)
printThread.Abort();
Thread.Abort();
Thread.Dispose();
That should do it if I'm not mistaken.
I have read that I can use asynchronous call with polling especially when the caller thread serves the GUI. I cannot see how because:
while(AsyncResult_.IsCompleted==false) //this stops the GUI thread
{
}
So how it come it should be good for this purpose? I needed to update my GUI status bar everytime deamon thread did some progress..
You are correct in your while loop stopping the GUI thread, when doing it like that, you don't want to do that.
If you need to poll, it would be better is to set up a Timer, and check whether the work has completed when the timer fires. The Timer can have a small resolution without problems (100 ms for instance), as long as you dont do much work during each tick.
However, I think you would be even better off by using a callback, so you do not need to poll and get notified as soon as your workload is done.
The point of async polling is that you can do other things in between checking IsCompleted — such as servicing GUI events. You could set a timer, for example, to trigger an event several times per second to check whether your asynchronous operation is finished, and use the normal GUI event loop to service those events together with all the other events your GUI receives. That way, your GUI remains responsive, and shortly after the async operation finishes, your timer event handler will notice it.
I was having the same trouble with an old API exposing BeginExecute() and EndExecute(). BeginExecute() started asynchrounous operation and then went silent until it finished executing to the end. But I was needed to update intermediate state of the execution progress in real-time.
So I came up with the following solution:
var asyncResult = command.BeginExecute();
while (!asyncResult.IsCompleted)
{
if (command.State != OldState)
{
progress.Report(newState);
}
// Key piece in this polling loop.
await Dispatcher.Yield(DispatcherPriority.ApplicationIdle);
}
command.EndExecute(asyncResult);
At first I have used
await Task.Yield();
But then I found out that in WPF it won't return the control to GUI, because this loop will have higher priority. That is why I switched to this instruction:
await Dispatcher.Yield(DispatcherPriority.ApplicationIdle);
So now GUI will check and update progress only when it has nothing else to do :)
I'm writing a plug-in for another program which uses the native program to open a series of files to extract some data from. One problem I am having is the process takes a long time and I want to keep the user interface from hanging. Plus I also want to give the user the ability to cancel the process before it completes. In the past I've used a background worker for this type of thing, but in this case I don't think a BackgroundWorker will work.
To create a plug-in through the API I am using one can create a custom command by inheriting from an IAPICommand interface. This interface includes an Execute(Application app) method. The class is then instantiated and the Execute() method is called by the program when the user evokes the custom command in the program.
The Execute() method is passed a reference to the current Application object when it is called, and it is this application object that is used to open the files to extract data from. However, the application instance is not able to open a document when requested by a thread other the the original Execute() thread.
So typically the UI would exist on the main thread, and the time consuming data extraction would be performed on a secondary thread. However, in this case the data extraction must be performed on the main thread, and I need to create a secondary thread for the UI.
Here's a stripped down version of the code.
class MyCommand:IAPICommand
{
public void Execute(Application app) // method from IAPICommand
{
Thread threadTwo= new Thread(ShowFormMethod);
threadTwo.Start();
}
public void ProcessWidget(Widget w, Application app)
{
//uses an App to work some magic on C
//app must be called from the original thread that called ExecuteCommand()
}
//method to open custom form on a seperatethread
public void ShowFormMethod()
{
MyForm form = new MyForm();
form.ShowDialog();
}
}
Here is a flow chart that shows how I think this should ultimately work.
alt text http://dl.dropbox.com/u/113068/SOMLibThreadingDiagram.jpg
Does this diagram make any sense, and if so am I even taking the correct approach to solve this problem?
Once the main thread starts the UI thread I want it to wait for the user to either select widgets to process, or end the command by closing the form (the red figures on the diagram). How can I make the main thread wait, and how do I trigger it to continue either with processing or to continue to the end when the UI thread ends? I was thinking I could have the main thread wait on a Monitor lock. The UI thread would then populate a static list of Widgets to be processed, and then pulse the main thread to trigger the processing. The UI thread would also pulse the Main thread when the form is closed, and the main thread would know to continue to the end of the command if it was ever pulsed when the list of widgets to process was empty.
How do I allow the main thread to communicate the progress or completion of widget processing back to the UI thread (yellow arrows in the diagram)? Do I just used the BeginInvoke() method of the Form to do this?
How do I allow the UI thread to cancel the widget processing (green arrow in the diagram)? I think I could just setup a static Boolean flag that is checked before each widget is processed?
It's generally a bad idea to have multiple threads in your application that each create forms. It isn't impossible to make this work, but it's much harder than you think it will be because forms that are in a parent-child relationship send messages to each other, and when they do, the one sending the message blocks until the one receiving handles it.
Mix this in with the message passing or synchronization between threads that you are doing explicitly, and it's easy to end up with deadlocks. So, in general, you are better off making sure that you reserve your main thread for your user interface, and do all processing in other threads that have no UI.
If you conform to that design, then the background threads can use Control.BeginInvoke to pass messages to the UI thread without having to wait for the messages to be processed.
In addition to the other answers, I recommend that you use a callback method from ProcessWidget to pass progress back to the calling thread. To prematurely stop the worker thread, you can use the callback to return a halt signal to your worker thread if it updates the caller often enough. Or use a separate callback method to periodically check for go/no-go. Or set a (gasp!) global static flag that the worker periodically checks. Or call Thread.Abort on the worker thread and have it catch the ThreadAbortException to clean up any resources.
I assume that the host application is a WinForms app.
You need to save the SynchronizationContext from the original thread in your Execute method, then call its Send method to execute code on the host's UI thread.
For example:
class MyCommand:IAPICommand
{
SynchronzationContext hostContext;
public void Execute(Application app) // method from IAPICommand
{
hostContext = SynchronzationContext.Current;
Thread threadTwo = new Thread(ShowFormMethod);
threadTwo.Start();
}
public void ProcessWidget(Widget w, Application app)
{
//uses an App to work some magic on C
//app must be called from the original thread that called ExecuteCommand()
SomeType someData = null;
hostContext.Send(delegate { someData = app.SomeMethod(); }, null);
}
}
If you look at Java swing, it is a nice example of how to do this:
1) A main thread is responsible for handling all UI requests. This removes any race conditions from the app.
2) Any time any "work" is to be done, spawn a thread (or a thread pool) and do the work. Thus the main thread is not held up except for a few microseconds and the UI is completely responsive while whatever is going on.
3) In all languages there has to be a thread interrupt mechanism. In java you invoke .interrupt() on the thread, and the current running thread gets a InterruptedException thrown wherever it is executing. You job is to catch that exception, figure out if you are really interrupted (read javadocs for this part) and if you are just let yourself die (return out of the run method).
1 + 2 = unobtrusive client interaction
3 = killing threads
An alternative to 3 (if 3 is too complex) is to give the thread a method .kill(); the method sets a kill flag. When you are reading a buffer from the hard drive in a loop, check if the kill flag is set, if it is then break out of the loop, close handlers, and return out of the run method.
Edit: sorry forgot to mention progress report:
Your thread should have a publicly exposed thread-safe method of getting the "progress report" or rather a data structure containing information about progress. Your UI thread should periodically (say every .5 seconds) check the thread's progress report and update the UI's progress bar. And by UI thread checking I mean your widget that shows the progress makes a request to re-render with the latest information on a timer, until done.