I'm using service remoting to connect service A with service B. While deploying, I've noticed that it is possible for service A to call a method in service B while service B is still starting up. In which case AddStringToDictionary throws a NullReferenceException because RunAsync has not been called yet.
public interface IServiceB : IService
{
Task AddStringToDictionary(string key, string value);
}
internal sealed class ServiceB : StatefulService, IServiceB
{
private IReliableDictionary<string, string> myDictionary;
public ServiceB(StatefulServiceContext context)
: base(context)
{
// StateMananger is null here.
}
public async Task AddStringToDictionary(string key, string value)
{
using (var tx = StateManager.CreateTransaction())
{
// myDictionary is null if this is called before RunAsync.
await this.myDictionary.SetAsync(tx, key, value);
await tx.CommitAsync();
}
}
protected override async Task RunAsync(CancellationToken cancellationToken)
{
this.myDictionary = await this.StateManager.GetOrAddAsync<IReliableDictionary<string, string>>("myDictionary");
}
}
I know that I can use the StateManager.GetOrAddAsync method in AddStringToDictionary, but I might want to initialize some more fields. The constructor is not useful either because the StateManager doesn't exist there yet. My current solution is to add a field:
private bool serviceAvailable = false;
I'm setting this field to true at the end of RunAsync. Every public method then calls ThrowIfServiceUnavailable().
private void ThrowIfServiceUnavailable()
{
if (!serviceAvailable)
{
throw new ServiceUnavailableException();
}
}
public async Task AddStringToDictionary(string key, string value)
{
ThrowIfServiceUnavailable();
...
}
Currently I have to handle the ServiceUnavailableException every time I call a method with remoting. My question is: are there any better ways to handle this situation or is this a good solution? And is it possible to add my custom exception to the ServiceProxy's transient exceptions to retry?
Here's an implementation that blocks calls while the service is being initialized. It uses ManualResetEventSlim to block.
And it uses SemaphoreSlim to enforce single threaded access to an async operation.
All Service operations call await WaitForInitializeAsync(CancellationToken.None);
And the implementation of this method is:
private async Task WaitForInitializeAsync(CancellationToken cancellationToken)
{
if (_initializer.IsSet) return;
await Task.Run(() => InitializeAsync(cancellationToken), cancellationToken);
_initializer.Wait(cancellationToken);
}
The implementation of InitializeAsync:
private async Task InitializeAsync(CancellationToken cancellationToken)
{
if (_initializer.IsSet) return;
try
{
_semaphore.Wait(cancellationToken);
if (_initializer.IsSet) return;
[initializer logic here]
Related
I am creating a Xamarin Forms application, and I am using the Xamarin Profiler to show that I have a memory leak. I have tracked the memory leak down to where it is happening, but I can't understand WHY it is happening.
I have a class (we will call it MyClass for now). And that class is using a Timer to call a service once every second. That service makes a REST call to retrieve a bunch of information, and then serializes the results back into an object....
MyClass:
public class MyClass : ContentPage
{
private readonly IMyService myService;
public MyClass() : base()
{
}
protected override async void OnAppearing()
{
StartTimer();
}
private void StartTimer()
{
Task.Run(async() =>
{
while(true)
{
myService = ((App)App.Current)
.serviceProvider
.GetRequiredService<IMyService>();
//--- everytime I call myService.GetSystemStatus(), my allocated memory continues to rise
MyResponse response = await myService.GetSystemStatus();
Device.BeginInvokeOnMainThread(() =>
{
// update the UI here...
});
await Task.Delay(1000);
}
});
}
}
MyService (Singleton):
public class MyService : IMyService
{
private readonly IMyHttpClientFactory httpClientFactory;
public MyService(IMyHttpClientFactory httpClientFactory)
{
this.httpClientFactory = httpClientFactory;
}
public async Task<MyResponse> GetSystemStatus()
{
return await httpClientFactory.Create().GetAsync<MyResponse>(
"http://example.com/api/status"
);
}
}
MyHttpClientFactory (Singleton):
public class MyHttpClientFactory : IMyHttpClientFactory
{
private readonly IServiceProvider _serviceProvider;
public MyHttpClientFactory(IServiceProvider serviceProvider)
{
_serviceProvider = serviceProvider;
}
public MyHttpClient Create()
{
return _serviceProvider.GetRequiredService<MyHttpClient>();
}
}
MyHttpClient:
public class MyHttpClient : IDisposable
{
private HttpClient _httpClient;
public MyHttpClient ()
{
_httpClient = new HttpClient();
_httpClient.Timeout = TimeSpan.FromSeconds(10);
}
public async Task<T> GetAsync<T>(string url) where T : new()
{
string s = await GetStringAsync(url);
return JsonConvert.DeserializeObject<T>(s);
}
public async Task<string> GetStringAsync(string url)
{
using (var response = await _httpClient.GetAsync(url))
{
response.EnsureSuccessStatusCode();
return await response.Content.ReadAsStringAsync();
}
}
}
My services are defined as follows:
public partial class App : Application
public ServiceProvider serviceProvider;
public App()
{
IServiceCollection services = new ServiceCollection();
ConfigureServices(services);
serviceProvider = services.BuildServiceProvider();
InitializeComponent();
}
private void ConfigureServices(IServiceCollection services)
{
services.AddHttpClient<MyHttpClient>("MyHttpClient", x =>
{
x.Timeout = TimeSpan.FromSeconds(5);
});
services.AddSingleton<IMyHttpClientFactory, MyHttpClientFactory>();
services.AddSingleton<IMyService, MyService>();
}
}
Best I can tell, the memory is going up because I am referencing the DI MyService inside a separate thread. But I am not sure if this is the reason or if there is something else that would be causing the leak?
Any advice would be greatly appreciated!!!
Thanks!
From what I understand from your code and your comments, it looks like you're looping by calling StartTimer() inside the Device.StartTimer() method.
According to the documentation, Device.StartTimer() is recurring and will occur every X seconds, depending of your interval parameter.
By removing the call to StartTimer() (the one between t.Dispose() and return false of MyClass.StartTimer, your code should work as expected and you will not create a new timer every x seconds
What could be the cause of the leak:
Your MyHttpClient class implements the IDisposable interface, yet the code to use an instance of this class is not leveraging the disposable nature of the object.
Even though the internal HttpClient instance is wrapped in a using statement, the MyHttpClient instance will not be disposed of as you would expect.
// from MyHttpClient class
public async Task<MyResponse> GetSystemStatus()
{
// no using statement here
return await httpClientFactory.Create().GetAsync<MyResponse>(
"http://example.com/api/status"
);
}
// should be:
public async Task<MyResponse> GetSystemStatus()
{
using (var client = await httpClientFactory.Create())
{
return await client.GetAsync<MyResponse>("http://example.com/api/status");
}
}
Another thing to try is to change the location of the resolution of the MyService instance to inside the Task since this is where it is used. This will allow the task to own the resource, and allow it to be collected when the task is complete.
private void StartTimer()
{
Device.StartTimer(TimeSpan.FromSeconds(1), () =>
{
Task t = Task.Run(async() =>
{
// resolve the service here
myService = ((App)App.Current)
.serviceProvider
.GetRequiredService<IMyService>();
MyResponse response = await myService.GetSystemStatus();
Device.BeginInvokeOnMainThread(() =>
{
// update the UI here...
});
});
t.Wait();
t.Dispose();
StartTimer();
return false;
});
}
A couple of additional observations of your code:
In your HttpClientFactory's Create() method, you are resolving an instance of your client from the DI container.
Your MyHttpClient class has a default constructor which means the resolution is not needed since there are no additional dependencies requiring DI support.
Your code could simply return a new MyHttpClient() instance from the Create() method without the need for DI.
Your MyHttpClient also implements the IMyHttpClient interface, but your factory returns the concrete type. This means you need to either remove the interface as unnecessary or change the return type to be the interface type since the interface is redundant unless it is used.
Thank you all for your answers....
I finally figured out the source of the memory leak.
The problem was that I was referencing "MyService" like this:
myService = ((App)App.Current)
.serviceProvider
.GetRequiredService<IMyService>();
The problem was that the serviceProvider object was a public property on my App. So each time I referenced the provider inside my loop, it was creating the leak.
To get around this, I added an abstract method to each of my pages that implemented MyClass to return the service correctly using DI. This has corrected my memory leak issue....
Thanks all for the help!
I don't think that your timer logic is the cause of the leak.
But in case it is useful to you, here is a clean way to do work periodically, yet if work takes a long time, avoid events "piling up".
Given await/async, no Timer is needed.
(There is an alternative solution that starts/stops a single System.Timers.Timer, but I won't go into that here.)
Replace StartTimer() declaration with the following:
/// <summary> Runs until keepRunning() returns false.
/// Delays by "msecDelay" AFTER finishing the previous loop's non-UI work. </summary>
private void StartTaskLoopWhileKeepRunning(Func<bool> keepRunning, int msecDelay = 250)
{
Task.Run(async () =>
{
while (keepRunning())
{
// Do non-UI work here.
// ... possibly slow work ...
Device.BeginInvokeOnMainThread(() =>
{
// NOTE: This work will run in parallel with the next delay.
// ... Do UI work here. ...
});
// Non-UI thread sleeps for msec.
await Task.Delay(msecDelay);
}
});
}
I am trying to implement a separate service for running a heavy time taking method run in background. I am kind of trying to do something like Hangfire but I can't use Hangfire because of some reasons.
So I have implement a package which takes in the delegate of the method and queues it and run it in background and also retries for some count if any exceptions. So I want to add one more functionality to it is if while retrying application is stop. It should continue it's from that count.
For above this to work I need to save this Func<CancellationToken, Task> delegate to DB. So that whenever again when we start I can get it from DB and run with all the details like exact arguments value.
Below is the code for better understanding and it is to understand the flow it does not contain retry logic.
This is BackgroundTaskQueue.cs
public class BackgroundTaskQueue : IBackgroundTaskQueue
{
private ConcurrentQueue<Func<CancellationToken, Task>> _workItems =
new ConcurrentQueue<Func<CancellationToken, Task>>();
private SemaphoreSlim _signal = new SemaphoreSlim(0);
private readonly ILogger<BackgroundTaskQueue> _logger;
public async Task QueueBackgroundWorkItemAsync(
Func<CancellationToken, Task> workItem, string jobId)
{
try{
if (workItem == null)
{
throw new ArgumentNullException(nameof(workItem));
}
_workItems.Enqueue(workItem);
_signal.Release();
}
catch (Exception ex) {
_logger.LogError(ex.Message);
}
}
public async Task<Func<CancellationToken, Task>> DequeueAsync(
CancellationToken cancellationToken)
{
await _signal.WaitAsync(cancellationToken);
_workItems.TryDequeue(out var workItem);
return workItem;
}
}
QueuedHostedService.cs
public class QueuedHostedService : BackgroundService
{
private readonly ILogger _logger;
private readonly IApplicationDbContext _dbContext;
private string jobId;
private Func<CancellationToken, Task>? workItem;
private CancellationToken cancellationToken;
public QueuedHostedService(IApplicationDbContext dbContext,
IBackgroundTaskQueue taskQueue,
ILoggerFactory loggerFactory)
{
TaskQueue = taskQueue;
_logger = loggerFactory.CreateLogger<QueuedHostedService>();
_dbContext = dbContext;
}
public IBackgroundTaskQueue TaskQueue { get; }
protected async override Task ExecuteAsync(
CancellationToken cancellationToken)
{
_logger.LogInformation("Queued Hosted Service is starting.");
this.cancellationToken = cancellationToken;
while (!this.cancellationToken.IsCancellationRequested)
{
workItem = await TaskQueue.DequeueAsync(cancellationToken);
await workItem(this.cancellationToken);
}
_logger.LogInformation("Queued Hosted Service is stopping.");
}
}
To make any method run in background we simply have to inject IBackgroundTaskQueue and call the method QueueBackgroundWorkItemAsync like below.
_queue.QueueBackgroundWorkItemAsync(async token => SomeMethod(arg1, arg2));
I want to make it same generic as Hangfire but I am unable to figure out what all fields I need to save to db from the Func<CancellationToken, Task>() and how do I get those fields. And another thing is how do I construct back that same Func<CancellationToken, Task>().
For above this to work I need to save this Func<CancellationToken, Task> delegate to DB.
This is how Hangfire works. It is one of the trickiest parts, and (IMO) it is also the worst part of the design. In particular, handling application upgrades can be tricky. It wasn't that long ago that Hangfire just flat-out didn't support rolling upgrades at all - now they do, but because of the delegate-serialization logic, they can only handle some types of application updates.
I recommend instead of trying to save a delegate, you have your code save a logical representation of the work to do. E.g.:
public interface IBackgroundWork { Task ExecuteAsync(CancellationToken token); }
public sealed class SomeBackgroundWork : IBackgroundWork
{
private readonly string _arg1;
private readonly string _arg2;
public SomeBackgroundWork(string arg1, string arg2)
{
_arg1 = arg1;
_arg2 = arg2;
}
public Task ExecuteAsync(CancellationToken token) =>
SomeMethodAsync(_arg1, _arg2);
}
Then serialize it appropriately (i.e., with a type field).
The disadvantage of this approach is that you need to define actual type representations for all of your background work item types. The advantage is that it fully captures all the logic without being obfuscated by delegate serialization, and it also is fully upgrade-safe, including rolling upgrades.
In a lot of our services, the OnStart method calls a neverEnding async Task and .Wait() which does all the service work until it gets manually shut down. We do something like the following (and have been since before I came onto the scene).
MyServiceClass service;
protected override void OnStart(string[] args)
{
eventLog.WriteEntry("MyService service is starting up.");
service = new MyServiceClass(this.ServiceName);
}
...
public class MyServiceClass
{
public MyServiceClass() {
Task.Run(() => DoServiceWork()).Wait();
}
public async Task DoServiceWork() {
while (true)
{
await Task.Delay(1000);//pretend this is actual work being done
}
}
}
Is there any reason not to just store the task and move on? (assuming the main method has appropriate exception handling)
public class MyServiceClass
{
public Task taskToBeAwaitedOnServiceStop;
public MyServiceClass() {
taskToBeAwaitedOnServiceStop = Task.Run(() => DoServiceWork());
}
}
or just to ignore the task entirely
public class MyServiceClass
{
public MyServiceClass() {
_ = DoServiceWork();
}
}
I'm considering the options as we don't like having .waits floating around (and having less means less to check when we go hunting). I'm asking here to be certain my alternatives won't cause strange service logic if implemented.
I hope this isn't too stupid a question, thanks for any help :)
Is it possible to share a queue between 2 or more stateful services, or do I need to directly call it via tcp/http to put a message on its own internal queue?
For example; say I have my first service that puts an order on a queue based on a condition:
public sealed class Service1 : StatefulService
{
public Service1(StatefulServiceContext context, IReliableStateManagerReplica reliableStateManagerReplica)
: base(context, reliableStateManagerReplica)
{ }
protected override async Task RunAsync(CancellationToken cancellationToken)
{
var customerQueue = await this.StateManager.GetOrAddAsync<IReliableQueue<Order>>("orders");
while (true)
{
cancellationToken.ThrowIfCancellationRequested();
using (var tx = this.StateManager.CreateTransaction())
{
if (true /* some logic here */)
{
await customerQueue.EnqueueAsync(tx, new Order());
}
await tx.CommitAsync();
}
}
}
}
Then my second service reads from that queue and then continues the processing.
public sealed class Service2 : StatefulService
{
public Service2(StatefulServiceContext context, IReliableStateManagerReplica reliableStateManagerReplica)
: base(context, reliableStateManagerReplica)
{ }
protected override async Task RunAsync(CancellationToken cancellationToken)
{
var customerQueue = await this.StateManager.GetOrAddAsync<IReliableQueue<Order>>("orders");
while (true)
{
cancellationToken.ThrowIfCancellationRequested();
using (var tx = this.StateManager.CreateTransaction())
{
var value = await customerQueue.TryDequeueAsync(tx);
if (value.HasValue)
{
// Continue processing the order.
}
await tx.CommitAsync();
}
}
}
}
I can't see much within the documentation on this, I can see that GetOrAddAsync method can take in a uri but I've seen no examples on how this works or if you can even do cross services?
The idea behind this is to split up the processing on to separate queues so that we don't get in a inconsistent state when we try to re-try a message.
There's no way to share state across services. The statemanager acts on a service partition level.
You could use an external queue for this purpose, like Service Bus.
You could also invert control, by using an Event Driven approach. Service 1 would raise an event, that Service 2 would use as a trigger to continue processing. The data to process could be inside the event, or data stored in another location, referenced to from the event.
In our application_startup, we seed up our database with some fake data, if no data exists.
To do this, we're using the Async methods to store the data. Great. Only problem is, we're not sure how to do this in the application_startup because that's not an async method.
I've spent soooo much time trying to understand #StevenCleary's tutorials and I'm always getting deadlocks. I totally grok what he consistently says:
As a general rule, you should use "async all the way down"; that is, don't block on async code
but I just don't get how I can do that, in this case :(
Lets imagine this is the code I'm trying to play with...
protected void Application_Start()
{
var someFakeData = LoadSomeFakeData();
var documentStore = new DocumentStore();
await documentStore.InitializeAsync(someFakeData);
...
// Registers this database as a singleton.
Container.Register(documentStore);
}
and later on .. some code that uses this documentStore. It is injected via construction injection ...
public SomeController(IDocumentStore documentStore)
{
_documentStore = documentStore;
}
public ViewModel GetFoos()
{
using (var session = _documentStore.OpenSession())
{
... db code goes in here ...
}
}
Clarification
I'm not trying to do some async code in here. I'm actually trying to call this async method, synchronously. Sure, i loose the benefits of async blah blah de blah.. but i'm happy with that. This is start up and I'm happy to block on startup.
In this case, you're asynchronously initializing a shared resource. So, I recommend that you either save the Task itself, or introduce an asynchronous wrapper type.
Using Task:
protected void Application_Start()
{
var someFakeData = LoadSomeFakeData();
var documentStore = new DocumentStore();
var documentStoreTask = documentStore.InitializeAsync(someFakeData);
...
// Registers this database task as a singleton.
Container.Register(documentStoreTask);
}
That may be too awkward, though, depending on Container. In that case, you can introduce an asynchronous wrapper type:
public sealed class DocumentStoreWrapper
{
private readonly Task<DocumentStore> _documentStore;
public DocumentStoreWrapper(Data data)
{
_documentStore = CreateDocumentStoreAsync(data);
}
private static async Task<DocumentStore> CreateDocumentStoreAsync(Data data)
{
var result = new DocumentStore();
await documentStore.InitializeAsync(data);
...
return result;
}
public Task<DocumentStore> DocumentStoreTask { get { return _documentStore; } }
}
protected void Application_Start()
{
var someFakeData = LoadSomeFakeData();
var documentStoreWrapper = new DocumentStoreWrapper(someFakeData);
...
// Registers this database wrapper as a singleton.
Container.Register(documentStoreWrapper);
}
Or, you could use AsyncLazy<T>, which does much the same thing but uses a background thread to execute the initialization code.
You can use of Task.Run(() => YourAsyncMethod()); inside of none async method like:
protected void Application_Start()
{
Task.Run(() => MyAsyncMethod(true));
}
This is an old topic, but it's popped up in my search and maybe it will for others.
For what the OP has requested (ie. To run an async method in a synchronous way from inside a synchronous method, and block until it's finished), is there some reason that the use of Task.WaitAll would not be a simple and adequate way of addressing this?
protected void Application_Start()
{
Task.WaitAll(MyAsyncMethod(true));
}
public static class AsyncHelper
{
private static readonly TaskFactory MyTaskFactory = new
TaskFactory(CancellationToken.None,
TaskCreationOptions.None,
TaskContinuationOptions.None,
TaskScheduler.Default);
public static TResult RunSync<TResult>(Func<Task<TResult>> func)
{
return MyTaskFactory
.StartNew(func)
.Unwrap()
.GetAwaiter()
.GetResult();
}
public static void RunSync(Func<Task> func)
{
MyTaskFactory
.StartNew(func)
.Unwrap()
.GetAwaiter()
.GetResult();
}
}
then use as
AsyncHelper.RunSync(ProcessAsync);
private async Task ProcessAsync(){ ....