I am trying to make a simple windows service that maintains a queue of integers and accepts new integers from other applications via a WCF call. My current implementation seems to maintain separate queues for each application which communicates with it, which is not what I want.
I started by following the instructions at from Microsoft on How to: Host a WCF Service in a Managed Windows Service.
My WindowsService class looks like this:
public class MyWindowsService : ServiceBase{
public ServiceHost serviceHost = null;
public MyWindowsService(){
ServiceName = "AdHocReportService";
}
public static void Main(){
ServiceBase.Run(new MyWindowsService());
}
protected override void OnStart(string[] args){
if (serviceHost != null)
serviceHost.Close();
serviceHost = new ServiceHost(typeof(MyService));
serviceHost.Open();
}
protected override void OnStop(){
if (serviceHost != null){
serviceHost.Close();
serviceHost = null;
}
}
}
In my Service class I have a queue and an Add method. The add method returns the count of items in the queue after the add. The code looks like this:
public class MyService : IMyService
{
private Queue<int> myQueue= new Queue<int>();
public int Add(int reportId)
{
myQueue.Enqueue(reportId);
return myQueue.Count;
}
}
Lastly, I test my service using the following code in a ConsoleApp:
MyServiceClient client = new MyServiceClient();
int count = client.Add(10);
Console.WriteLine(count); //prints 1
count = client.Add(25);
Console.WriteLine(count); //prints 2
Console.ReadLine();
I would expect this to print 1 and 2 the first time my test is run, then 3 and 4 the second time, and then 5 and 6 the third and so on. However, it simply returns 1 and 2 each time, as if the Console App is instantiating the object itself and not operating on the object inside the Windows Service. What am I not understanding?
I think you want a singleton WCF service. See here.
By default, the instance mode for a WCF service is per-call. So an instance of your service is being created by the host for each call you make.
Note: When using a singleton service, your operations need to be thread safe. So I suggest switching from a Queue to a ConcurrentQueue, so you can handle multiple concurrent clients.
Alternative: Use a MSMQ binding. This will ensure you that all of your incoming messages are queued out of process, therefore persisted between restarts too.
Related
I’ve created an object that I would like to pass in a WCF call… but inside ServiceReference1… this object is redefined… is there a way to just use the original object everywhere… it seems like people have done this but I can’t figure out what I am doing wrong.
The object is used as a parameter to a function in the service contract.
[OperationContract(IsOneWay = true)]
void UpdateInformation(MyObject myObject);
The error that I get when I try to call the function from my client is “Argument 1: cannot convert from ‘MyNameSpaceDTO.MyObject' to ‘MyNameSpace.ServiceReference1.MyObject’”
The object is in it’s own class library dll and it is marked with [DataObject] and [DataMember] attributes.
namespace MyNameSpaceDTO
{
[DataContract]
public class MyObject
{
[DataMember]
public string Name { get; set; }
….
But, also ends up in Reference.cs after adding the Service Reference as:
[System.Diagnostics.DebuggerStepThroughAttribute()]
[System.CodeDom.Compiler.GeneratedCodeAttribute("System.Runtime.Serialization", "4.0.0.0")]
[System.Runtime.Serialization.DataContractAttribute(Name="MyObject", Namespace="http://schemas.datacontract.org/2004/07/MyNameSpaceDTO")]
[System.SerializableAttribute()]
public partial class MyObject : object, System.Runtime.Serialization.IExtensibleDataObject, System.ComponentModel.INotifyPropertyChanged {
[System.NonSerializedAttribute()]
private System.Runtime.Serialization.ExtensionDataObject extensionDataField;
[System.Runtime.Serialization.OptionalFieldAttribute()]
private string NameField;
...
Also, I do have the following set in the Advanced section of the Add Service Reference:
[x] Reuse types in referenced assemblies
(o) Reuse types in all referenced assemblies
For consuming a WCF service you often see samples (and they're undoubtedly advisable!) where you're instructed to add that service via the Add Service Reference dialog. By referencing a service that way your client application creates proxy classes form the WSDL exposed by the service.
As a result you end up having e.g. a class MyNameSpaceDTO.MyObject in your contract-assembly and a MyNameSpace.ServiceReference1.MyObject in your client application which was generated form the WSDL. This may seem somewhat redundant.
One situation in which you may need this behaviour could be the following: Imagine you'd want to consume an arbitrary public web service which you don't control. You have no access to the contract-assembly which defines the types etc. In that situation creating your own local proxy classes from the exposed WSDL is optimal since it's your only way to get the needed types and so on.
But your concrete situation seems to be a little bit different. I think what you're looking for is a shared contract. Since you're in control of the client and server code (and both live happily side by side in the same solution), you're in the comfortable situation to just share the contract:
So instead of adding a service reference within your client-app (via Add Service Reference), you'd just reference the contract-assembly (via the usual Add Reference dialogue). By doing this there'll by only one MyNameSpaceDTO.MyObject since the second one is never created and not needed. This approach is called contract sharing.
Please take a look at that example:
EDIT:
Please note some changes: The most important one is that you usually wouldn't want to share the assembly which holds your implementation logic of your service. So I extracted that part from the Contract-assembly and put it in a separate Implementation-assembly. By doing so, you simply share the interfaces and types and not the implementation logic. This change is reflected in the screenshot above, too.
You could set up that small solution with the following classes:
Contract - IService1.cs:
[ServiceContract]
public interface IService1
{
[OperationContract]
string GetData(int value);
}
Implementation - Service1.cs:
public class Service1 : IService1
{
public string GetData(int value)
{
return string.Format("You entered: {0}", value);
}
}
Host - Program.cs:
class Program
{
static void Main(string[] args)
{
var baseAddress = new Uri("http://localhost:8732/Design_Time_Addresses/Service1/");
using (var host = new ServiceHost(typeof(Service1), baseAddress))
{
// Enable metadata publishing.
var smb = new ServiceMetadataBehavior();
smb.HttpGetEnabled = true;
smb.MetadataExporter.PolicyVersion = PolicyVersion.Policy15;
host.Description.Behaviors.Add(smb);
// Open the ServiceHost to start listening for messages. Since no endpoints are
// explicitly configured, the runtime will create one endpoint per base address
// for each service contract implemented by the service.
host.Open();
Console.WriteLine("The service is ready at {0}", baseAddress);
Console.WriteLine("Press <Enter> to stop the service.");
Console.ReadLine();
host.Close();
}
}
}
Client - Program.cs:
class Program
{
static void Main(string[] args)
{
Console.WriteLine("Press <Enter> to proceed.");
Console.ReadLine();
var binding = new BasicHttpBinding();
var endpoint = new EndpointAddress("http://localhost:8732/Design_Time_Addresses/Service1/");
var channelFactory = new ChannelFactory<IService1>(binding, endpoint);
// Create a channel.
IService1 wcfClient1 = channelFactory.CreateChannel();
string s = wcfClient1.GetData(42);
Console.WriteLine(s);
((IClientChannel)wcfClient1).Close();
Console.WriteLine("Press <Enter> to quit the client.");
Console.ReadLine();
}
}
My objective is to implement an asynchronous self hosted WCF service which will run all requests in a single thread and make full use of the new C# 5 async features.
My server will be a Console app, in which I will setup a SingleThreadSynchronizationContext, as specified here, create and open a ServiceHost and then run the SynchronizationContext, so all the WCF requests are handled in the same thread.
The problem is that, though the server was able to successfully handle all requests in the same thread, async operations are blocking the execution and being serialized, instead of being interlaced.
I prepared a simplified sample that reproduces the issue.
Here is my service contract (the same for server and client):
[ServiceContract]
public interface IMessageService
{
[OperationContract]
Task<bool> Post(String message);
}
The service implementation is the following (it is a bit simplified, but the final implementation may access databases or even call other services in asynchronous fashion):
public class MessageService : IMessageService
{
public async Task<bool> Post(string message)
{
Console.WriteLine(string.Format("[Thread {0} start] {1}", Thread.CurrentThread.ManagedThreadId, message));
await Task.Delay(5000);
Console.WriteLine(string.Format("[Thread {0} end] {1}", Thread.CurrentThread.ManagedThreadId, message));
return true;
}
}
The service is hosted in a Console application:
static void Main(string[] args)
{
var syncCtx = new SingleThreadSynchronizationContext();
SynchronizationContext.SetSynchronizationContext(syncCtx);
using (ServiceHost serviceHost = new ServiceHost(typeof(MessageService)))
{
NetNamedPipeBinding binding = new NetNamedPipeBinding(NetNamedPipeSecurityMode.None);
serviceHost.AddServiceEndpoint(typeof(IMessageService), binding, address);
serviceHost.Open();
syncCtx.Run();
serviceHost.Close();
}
}
As you can see, the first thing I do is to setup a single threaded SynchronizationContext. Following, I create, configure and open a ServiceHost. According to this article, as I've set the SynchronizationContext prior to its creation, the ServiceHost will capture it and all the client requests will be posted in the SynchronizationContext. In the sequence, I start the SingleThreadSynchronizationContext in the same thread.
I created a test client that will call the server in a fire-and-forget fashion.
static void Main(string[] args)
{
EndpointAddress ep = new EndpointAddress(address);
NetNamedPipeBinding binding = new NetNamedPipeBinding(NetNamedPipeSecurityMode.None);
IMessageService channel = ChannelFactory<IMessageService>.CreateChannel(binding, ep);
using (channel as IDisposable)
{
while (true)
{
string message = Console.ReadLine();
channel.Post(message);
}
}
}
When I execute the example, I get the following results:
Client
Server
The messages are sent by the client with a minimal interval ( < 1s).
I expected the server would receive the first call and run it in the SingleThreadSynchronizationContext (queueing a new WorkItem. When the await keyword was reached, the SynchronizationContext would be once again captured, the continuation posted to it, and the method would return a Task at this point, freeing the SynchronizationContext to deal with the second request (at least start dealing with it).
As you can see by the Thread's id in the server log, the requests are being correctly posted in the SynchronizationContext. However, looking at the timestamps, we can see that the first request is being completed before the second is started, what totally defeats the purpose of having a async server.
Why is that happening?
What is the correct way of implementing a WCF self hosted async server?
I think the problem is with the SingleThreadSynchronizationContext, but I can't see how to implement it in any other manner.
I researched the subject, but I could not find more useful information on asynchronous WCF service hosting, especially using the Task based pattern.
ADDITION
Here is my implementation of the SingleThreadedSinchronizationContext. It is basically the same as the one in the article:
public sealed class SingleThreadSynchronizationContext
: SynchronizationContext
{
private readonly BlockingCollection<WorkItem> queue = new BlockingCollection<WorkItem>();
public override void Post(SendOrPostCallback d, object state)
{
this.queue.Add(new WorkItem(d, state));
}
public void Complete() {
this.queue.CompleteAdding();
}
public void Run(CancellationToken cancellation = default(CancellationToken))
{
WorkItem workItem;
while (this.queue.TryTake(out workItem, Timeout.Infinite, cancellation))
workItem.Action(workItem.State);
}
}
public class WorkItem
{
public SendOrPostCallback Action { get; set; }
public object State { get; set; }
public WorkItem(SendOrPostCallback action, object state)
{
this.Action = action;
this.State = state;
}
}
You need to apply ConcurrencyMode.Multiple.
This is where the terminology gets a bit confusing, because in this case it doesn't actually mean "multi-threaded" as the MSDN docs state. It means concurrent. By default (single concurrency), WCF will delay other requests until the original operation has completed, so you need to specify multiple concurrency to permit overlapping (concurrent) requests. Your SynchronizationContext will still guarantee only a single thread will process all the requests, so it's not actually multi-threading. It's single-threaded concurrency.
On a side note, you might want to consider a different SynchronizationContext that has cleaner shutdown semantics. The SingleThreadSynchronizationContext you are currently using will "clamp shut" if you call Complete; any async methods that are in an await are just never resumed.
I have an AsyncContext type that has better support for clean shutdowns. If you install the Nito.AsyncEx NuGet package, you can use server code like this:
static SynchronizationContext syncCtx;
static ServiceHost serviceHost;
static void Main(string[] args)
{
AsyncContext.Run(() =>
{
syncCtx = SynchronizationContext.Current;
syncCtx.OperationStarted();
serviceHost = new ServiceHost(typeof(MessageService));
Console.CancelKeyPress += Console_CancelKeyPress;
var binding = new NetNamedPipeBinding(NetNamedPipeSecurityMode.None);
serviceHost.AddServiceEndpoint(typeof(IMessageService), binding, address);
serviceHost.Open();
});
}
static void Console_CancelKeyPress(object sender, ConsoleCancelEventArgs e)
{
if (serviceHost != null)
{
serviceHost.BeginClose(_ => syncCtx.OperationCompleted(), null);
serviceHost = null;
}
if (e.SpecialKey == ConsoleSpecialKey.ControlC)
e.Cancel = true;
}
This will translate Ctrl-C into a "soft" exit, meaning the application will continue running as long as there are client connections (or until the "close" times out). During the close, existing client connections can make new requests, but new client connections will be rejected.
Ctrl-Break is still a "hard" exit; there's nothing you can do to change that in a Console host.
I have a windows service and a GUI that need to communicate with each other. Either can send messages at any time.
I'm looking at using NamedPipes, but it seems that you cant read & write to the stream at the same time (or at least I cant find any examples that cover this case).
Is it possible to do this kind of two-way communication via a single NamedPipe?
Or do I need to open two pipes (one from GUI->service and one from service->GUI)?
Using WCF you can use duplex named pipes
// Create a contract that can be used as a callback
public interface IMyCallbackService
{
[OperationContract(IsOneWay = true)]
void NotifyClient();
}
// Define your service contract and specify the callback contract
[ServiceContract(CallbackContract = typeof(IMyCallbackService))]
public interface ISimpleService
{
[OperationContract]
string ProcessData();
}
Implement the Service
[ServiceBehavior(InstanceContextMode=InstanceContextMode.PerCall)]
public class SimpleService : ISimpleService
{
public string ProcessData()
{
// Get a handle to the call back channel
var callback = OperationContext.Current.GetCallbackChannel<IMyCallbackService>();
callback.NotifyClient();
return DateTime.Now.ToString();
}
}
Host the Service
class Server
{
static void Main(string[] args)
{
// Create a service host with an named pipe endpoint
using (var host = new ServiceHost(typeof(SimpleService), new Uri("net.pipe://localhost")))
{
host.AddServiceEndpoint(typeof(ISimpleService), new NetNamedPipeBinding(), "SimpleService");
host.Open();
Console.WriteLine("Simple Service Running...");
Console.ReadLine();
host.Close();
}
}
}
Create the client application, in this example the Client class implements the call back contract.
class Client : IMyCallbackService
{
static void Main(string[] args)
{
new Client().Run();
}
public void Run()
{
// Consume the service
var factory = new DuplexChannelFactory<ISimpleService>(new InstanceContext(this), new NetNamedPipeBinding(), new EndpointAddress("net.pipe://localhost/SimpleService"));
var proxy = factory.CreateChannel();
Console.WriteLine(proxy.ProcessData());
}
public void NotifyClient()
{
Console.WriteLine("Notification from Server");
}
}
Using a single point to accumulate messages (a single pipe in this case) forces you to handle direction of the message yourself too (in addition to that you have to use a system-wide lock for the pipe).
So use 2 pipes with opposite directions.
(Another option would be using 2 MSMQ queues).
Your named pipe stream classes (server or client) have to be constructed with a PipeDirection of InOut. You need one NamedPipeServerStream, probably in your service, which can be shared by an arbitrary number of NamedPipeClientStream objects. Construct the NamedPipeServerStream with the name of the pipe and the direction, and the NamedPipeClientStream with the name of the pipe, the name of the server, and the PipeDirection, and you should be good to go.
if I use this: InstanceContextMode.PerSession, then for every client I initialize my service one time. Am I right?
If I call a method that have [OperationBehavior(TransactionScopeRequired = true)], JobImplement constructor gets called. Why?
My Service:
[ServiceBehavior(InstanceContextMode = InstanceContextMode.PerSession)]
public class JobImplement : IJob
{
public static int Sum = 0;
public JobImplement()
{
Sum++;
}
[OperationBehavior(TransactionScopeRequired = true)]
public void UselessMethod1()
{
}
[OperationBehavior(TransactionScopeRequired = true)]
public void AddEmployee(string Name, string Age)
{
}
public int GetSum()
{
return Sum;
}
}
EDIT 1:
Im using WSHttpBinding binding and here is my only client code.
My client:
static void Main(string[] args)
{
ServiceReference1.IJob Service1 = new ServiceReference1.JobClient();
Service1.UselessMethod1();//sum become [1]
Service1.AddEmployee("","");//sum become [2]
Console.WriteLine(Service1.GetSum());//show [2].
Console.WriteLine(Service1.GetSum());//show [2].
Console.WriteLine(Service1.GetSum());//show [2].
}
The service as written here uses the default value of the ServiceBehavior.ReleaseServiceInstanceOnTransactionComplete attribute, which is true.
This means that the methods enforcing transaction scope via TransactionScopeRequired will cause the service instance to be released and the next time a method is invoked, a new instance gets created despite your desired InstanceContextMode.
Changing the relevant attribute to false should solve the problem.
Looks like your new instance of the service created every time client calls your service. It might be because client terminated session (and your constructor). For instance client closed proxy, in this case next time client calls service, new session will be created. Did you check you are not closing proxy every time you connect to the service?
I'm new with WCF and still experimenting.
I have two ServiceContract like this:
[ServiceContract]
public interface IFirst
{
[OperationContract(IsOneWay = true)]
void First();
}
[ServiceContract]
public interface ISecond
{
[OperationContract(IsOneWay = true)]
void Second();
}
On server side Client class implementing those interfaces:
public class Client : IFirst, ISecond
{
static int count = 0;
int id;
public Client()
{
count++;
id = count;
Console.WriteLine("{0} client created.", id);
}
public void First()
{
Console.WriteLine("First from: " + id.ToString());
}
public void Second()
{
Console.WriteLine("Second: " + id.ToString());
}
}
and host:
ServiceHost host = new ServiceHost(typeof(Client), new Uri("net.tcp://localhost:8000"));
NetTcpBinding binding = new NetTcpBinding();
host.AddServiceEndpoint(typeof(IFirst), binding, "");
host.AddServiceEndpoint(typeof(ISecond), binding, "");
host.Open();
On client side:
ChannelFactory<IFirst> firstFactory = new ChannelFactory<IFirst>(new NetTcpBinding(), new EndpointAddress("net.tcp://localhost:8000"));
IFirst iFirst = firstFactory.CreateChannel();
iFirst.First();
ChannelFactory<ISecond> secondFactory = new ChannelFactory<ISecond>(new NetTcpBinding(), new EndpointAddress("net.tcp://localhost:8000"));
ISecond iSecond = secondFactory.CreateChannel();
iSecond.First();
This works fine. It calls First and then Second method but for each call creates new instance of Client class. It is obvious because when client creates channel, service creates new instance of Client object. What I wish to achieve is call Second after First but for the same instance of Client. Is it even possible? I know I can put this two methods in one service but it is not what I wish for.
As per my understanding you want to call the two methods while creating only one client.
In WCF you can control Instancing by setting InstanceContextMode property of Service Behavior Attribute. There are three possible values
-PerCall
-PerSession
-Single
You can use PerSession as it keeps the object active for the next calls from client.
The object is released when the session ends
You can decorate your class
[ServiceBehavior(InstanceContextMode=InstanceContextMode.PerSession)]
Hope this helps.
At first I was a little confused by you calling your Service Implementation "Client" :P
But anyway, by default, WCF uses InstanceContextMode.PerCall, which means that it will instantiate a new Service implementation instance for every request into the Service.
If you want for subsequent service calls to be handled within the same service instance, you will have to use either PerSession or Single InstanceContextModes.
http://msdn.microsoft.com/en-us/library/ms731193.aspx summarises Sessions, Instances and Concurrency in WCF pretty well.