hazelcast / hazelcast-csharp-client

Hazelcast .NET Client

.NET Client

Hazelcast .NET brings the full power of the Hazelcast high-performance, in-memory computing platform to the Microsoft .NET ecosystem. The Hazelcast .NET Client allows you to elastically scale your .NET caches at high read speeds, to access all of Hazelcast data structures such as distributed maps, queues, topics and more. All, with enterprise level security through SSL and mutual authentication.

Documentation for the client is provided on the documentation site, with examples, guides and FAQs, and a complete reference documentation for the public API.

The .NET client itself is distributed via NuGet as a package named Hazelcast.NET. It can be installed like any other NuGet package, either via the Visual Studio GUI, or via the package manager. Note that Hazelcast is composed of two parts: the server, and the client. Browse to Hazelcast In-Memory Computing to find out how to obtain and run the server part.

The Hazelcast .NET solution is Open Source, released under the Apache License 2.0, and welcomes contributions. The project lives here on GitHub, where you can obtain the source code, report issues, and interract with the community. Contributions are welcome!

Versions

Browse to this page for details about versions.

See this branch for more information about version 3 of the client.

Code Samples

Check the Hazelcast.Net.Examples project.

Contributing

We encourage any type of contribution in the form of issue reports or pull requests.

Issue Reports

For issue reports, please share the following information with us to quickly resolve the problems.

• Hazelcast IMDG and the client version that you use
• General information about the environment and the architecture you use like Node.js version, cluster size, number of clients, Java version, JVM parameters, operating system etc.
• Logs and stack traces, if any.
• Detailed description of the steps to reproduce the issue.

Pull Requests

Contributions are submitted, reviewed and accepted using the pull requests on GitHub. For an enhancement or larger feature, create a GitHub issue first to discuss.

Development

Development for future versions takes place in the master branch (this branch). The documentation site contains details about building, testing and running the code.

License

Apache 2.0 License.

Copyright

Copyright (c) 2008-2022, Hazelcast, Inc. All Rights Reserved.

Visit www.hazelcast.com for more information.

hazelcast-csharp-client's Issues

All hazelcast exception should support builtin .Net serialization.

Exception like HazelcastException can be marked as Serializable to support containers some remote calls etc.

Client Statistics

Client related info and statistics for Management Center.

• Client memory usage
• Client CPU Usage
• Client connected time
• Near cache statistics

Fast Aggregation and projection implementation

ClientRetryTest.ClientTransactionRetry

ESSAGE:
                                                An unexpected exception type was thrown
Expected: System.InvalidOperationException
 but was: Hazelcast.Transaction.TransactionException : No transaction context found for txnId:559caf37-ff27-4962-aa99-e83e36ed347b
                                                +++++++++++++++++++
                                                STACK TRACE:
                                                at Hazelcast.Client.Proxy.ClientTxnProxy.Invoke(IClientMessage request) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Proxy\ClientTxnProxy.cs:line 98
at Hazelcast.Client.Proxy.ClientTxnProxy.Invoke[T](IClientMessage request, Func`2 decodeResponse) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Proxy\ClientTxnProxy.cs:line 102
at Hazelcast.Client.Proxy.ClientTxnMapProxy`2.Put(TKey key, TValue value, Int64 ttl, TimeUnit timeunit) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Proxy\ClientTxnMapProxy.cs:line 77
at Hazelcast.Client.Proxy.ClientTxnMapProxy`2.Put(TKey key, TValue value) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Proxy\ClientTxnMapProxy.cs:line 67
at Hazelcast.Client.Test.ClientRetryTest.ClientTransactionRetry() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Test\Hazelcast.Client.Test\ClientRetryTest.cs:line 54

https://hazelcast-l337.ci.cloudbees.com/view/Official%20Builds/job/csharp-client-master/73/testReport/junit/Hazelcast.Client.Test/ClientRetryTest/ClientTransactionRetry/

Hanging on client creation

Hello
If I start a dotnet client prior to any hazelcast node, then start an hazelcast instance, the client connects but never completes the initialization and hangs forever.
The code used on the dotnet side:

        var config = new ClientConfig();
        config.GetNetworkConfig().AddAddress(new[] { "127.0.0.1:5701" });
        var client = HazelcastClient.NewHazelcastClient(config); // stay here for ever

To start an hazelcast instance I simply use the following command:

  start java -jar hazelcast-3.6-RC2-20151203.155725-88.jar

The logs that I get on the dotnet side:

22:19:03.772 [FINEST] - [:8] Hazelcast.Client.Spi.ClientClusterService: Trying to connect to Address[127.0.0.1]:5701
22:19:03.772 [FINEST] - [:8] Hazelcast.Client.Connection.IClientConnectionManager: Creating new connection for Address[127.0.0.1]:5701 with id 0
22:19:03.784 [FINEST] - [:8] Hazelcast.Client.Spi.ClientClusterService: Authenticating against the owner node
// nothing after that

Let me know if I have missed anything or if you need more information

Cluster member shows wrong version when the client is connected

When a Hazelcast .NET 3.8.1 client connected to a cluster, members reports the connectivity as follows:

INFO: [192.168.56.1]:5701 [dev] [3.8.1] Received auth from Connection[id=1, /127.0.0.1:5701->/127.0.0.1:11255, endpoint=null, alive=true, type=CSHARP_CLIENT], successfully authenticated, principal : ClientPrincipal{uuid='81d52973-8480-441d-a2b2-c0a8aeb49fe6', ownerUuid='455b9a86-453b-480c-acaa-025e54fab2f8'}, owner connection : true, client version : 3.8.0

As it is seen, the client version seems as 3.8.0 which is 3.8.1 in real.

ClientMapTest.TestAsyncPutWithTtl

MESSAGE:
Expected: "value1"
But was:  null
+++++++++++++++++++
STACK TRACE:
at Hazelcast.Client.Test.ClientMapTest.TestAsyncPutWithTtl() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Test\Hazelcast.Client.Test\ClientMapTest.cs:line 190

https://hazelcast-l337.ci.cloudbees.com/view/Official%20Builds/job/csharp-client-master/111/testReport/junit/Hazelcast.Client.Test/ClientMapTest/TestAsyncPutWithTtl/

HazelcastSerializationException is thrown when doing operations in parallel

When running a test which loads data in parallel, the following exception is thrown:

System.AggregateException : One or more errors occurred.

----> Hazelcast.IO.Serialization.HazelcastSerializationException : Class-id: 106 is already registered!

at System.Threading.Tasks.Task.ThrowIfExceptional(Boolean includeTaskCanceledExceptions)

at System.Threading.Tasks.Task.Wait(Int32 millisecondsTimeout, CancellationToken cancellationToken)

at System.Threading.Tasks.Parallel.ForWorker[TLocal](Int32 fromInclusive, Int32 toExclusive, ParallelOptions parallelOptions, Action`1 body, Action`2 bodyWithState, Func`4 bodyWithLocal, Func`1 localInit, Action`1 localFinally)

at System.Threading.Tasks.Parallel.ForEachWorker[TSource,TLocal](IEnumerable`1 source, ParallelOptions parallelOptions, Action`1 body, Action`2 bodyWithState, Action`3 bodyWithStateAndIndex, Func`4 bodyWithStateAndLocal, Func`5 bodyWithEverything, Func`1 localInit, Action`1 localFinally)

at System.Threading.Tasks.Parallel.ForEach[TSource](IEnumerable`1 source, Action`1 body)

--HazelcastSerializationException

at Hazelcast.IO.Serialization.SerializationService.ToObject[T](Object object)

at Hazelcast.Client.Proxy.ClientMapProxy`2.Get(Object key)

When it is changed the parallel client down to be single threaded this issue does not arise.

Implement versioned serialization read format for 3.8

ClientDistributedObjectTest.TestGetDistributedObjects

 MESSAGE:
                                                  Expected: 3
  But was:  4

                                                +++++++++++++++++++
                                                STACK TRACE:
                                                at Hazelcast.Client.Test.ClientDistributedObjectTest.TestGetDistributedObjects() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Test\Hazelcast.Client.Test\ClientDistributedObjectTest.cs:line 49

https://hazelcast-l337.ci.cloudbees.com/view/Official%20Builds/job/csharp-client-master/172/testReport/junit/Hazelcast.Client.Test/ClientDistributedObjectTest/TestGetDistributedObjects/

Error while compiling the Hazelcast.Test project

Hello,

I am trying to create hazelcast dot net client.
While compiling the test project I am getting the below error:

Error 13 Friend access was granted to 'Hazelcast.Test, PublicKey=00240000048000009400000006020000002400005253413100040000010001004d81045a994968ac643918d7bbce405b2473471d8de6aed6bbffc0fe1874bfcabf3c0b437c6c5293a589bdcbe884c6d86934069b35deaf5ab2e770cbff41a20dd4014bb53e481c30bd3ead29437b02dec5916a717a4a2b4fd353e81238b89ae09e5ba0ab615c5fef7937aabab4e240c3dffe2b948047769eeb07f674589d0bb3', but the output assembly is named 'Hazelcast.Test, Version=3.8.0.0, Culture=neutral, PublicKeyToken=null'. Try adding a reference to 'Hazelcast.Test, PublicKey=00240000048000009400000006020000002400005253413100040000010001004d81045a994968ac643918d7bbce405b2473471d8de6aed6bbffc0fe1874bfcabf3c0b437c6c5293a589bdcbe884c6d86934069b35deaf5ab2e770cbff41a20dd4014bb53e481c30bd3ead29437b02dec5916a717a4a2b4fd353e81238b89ae09e5ba0ab615c5fef7937aabab4e240c3dffe2b948047769eeb07f674589d0bb3' or changing the output assembly name to match. d:Hazelcast.Net\bin\Debug\Hazelcast.Net.dll Hazelcast.Test

Also, I want to connect to the Hazelcast cache server, which password protected. Please let me know how to do this.

Thanks.

Serialization service memory leak

Serialization service internal bufferPool leaks due to strong Thread references.

Reported by #115

Partition Predicate Implementation

fixed by #117

ReliableTopic [API-1958]

Is ReliableTopic in the plans for this client? If yes, approximately when? Cheers!

ClientQueueTest.TestAddAll

 MESSAGE:
                                                SetUp : Hazelcast.Core.HazelcastException : com.hazelcast.spi.exception.RetryableIOException: Packet not send to -> Address[172.30.0.171]:5702
  ----> Hazelcast.Core.HazelcastException : Exception of type 'Hazelcast.Core.HazelcastException' was thrown.
TearDown : Hazelcast.Core.HazelcastInstanceNotActiveException : Hazelcast instance is not active!
                                                +++++++++++++++++++
                                                STACK TRACE:
                                                at Hazelcast.Client.Spi.ProxyManager.Initialize(ClientProxy clientProxy) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Spi\ProxyManager.cs:line 287
at Hazelcast.Client.Spi.ProxyManager.InitializeWithRetry(ClientProxy clientProxy) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Spi\ProxyManager.cs:line 315
at Hazelcast.Client.Spi.ProxyManager.GetOrCreateProxy[T](String service, String id) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Spi\ProxyManager.cs:line 129
at Hazelcast.Client.HazelcastClient.GetDistributedObject[T](String serviceName, String name) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client\HazelcastClient.cs:line 236
at Hazelcast.Client.HazelcastClient.GetQueue[T](String name) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client\HazelcastClient.cs:line 115
at Hazelcast.Client.HazelcastClientProxy.GetQueue[T](String name) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client\HazelcastClientProxy.cs:line 40
at Hazelcast.Client.Test.ClientQueueTest.Init() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Test\Hazelcast.Client.Test\ClientQueueTest.cs:line 35
--HazelcastException

--TearDown
at Hazelcast.Client.Spi.ClientProxy.Invoke(IClientMessage request, Object key) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Spi\ClientProxy.cs:line 163
at Hazelcast.Client.Proxy.ClientQueueProxy`1.Invoke(IClientMessage request) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Proxy\ClientQueueProxy.cs:line 301
at Hazelcast.Client.Spi.ClientProxy.Invoke[T](IClientMessage request, Func`2 decodeResponse) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Spi\ClientProxy.cs:line 186
at Hazelcast.Client.Proxy.ClientQueueProxy`1.Clear() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Proxy\ClientQueueProxy.cs:line 285
at Hazelcast.Client.Test.ClientQueueTest.Destroy() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Test\Hazelcast.Client.Test\ClientQueueTest.cs:line 42

https://hazelcast-l337.ci.cloudbees.com/view/Official%20Builds/job/csharp-client-master/184/testReport/junit/Hazelcast.Client.Test/ClientQueueTest/TestAddAll/

Upgrade client protocol to 1.4

update with regenerated client protocol codecs

Decrement Pending packets for events also

An issue found in java client.
hazelcast/hazelcast#6608
Needs to be checked if same issue is here.

IHazelcastInstance does not implement IDisposable but has a Shutdown method

ClientSSLTest

Hazelcast.Client.Test.ClientSSLTest.TestMapSSLEnabled_DoNotValidateName

MESSAGE:
                                                System.InvalidOperationException : Unable to connect to any address in the config! The following addresses were tried:127.0.0.1:5701
                                                +++++++++++++++++++
                                                STACK TRACE:
                                                at Hazelcast.Client.Spi.ClientClusterService.ConnectToOne() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Spi\ClientClusterService.cs:line 379
at Hazelcast.Client.Spi.ClientClusterService.ConnectToCluster() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Spi\ClientClusterService.cs:line 333
at Hazelcast.Client.Spi.ClientClusterService.Start() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Spi\ClientClusterService.cs:line 225
at Hazelcast.Client.HazelcastClient.Start() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client\HazelcastClient.cs:line 427
at Hazelcast.Client.HazelcastClient.NewHazelcastClient(ClientConfig config) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client\HazelcastClient.cs:line 329
at Hazelcast.Client.Test.ClientSSLTest.Setup(Boolean ssl, Boolean validateCertificateChain, Boolean validateCertificateName, String certificateName) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Test\Hazelcast.Client.Test\ClientSSLTest.cs:line 40
at Hazelcast.Client.Test.ClientSSLTest.TestMapSSLEnabled_DoNotValidateName() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Test\Hazelcast.Client.Test\ClientSSLTest.cs:line 74

https://hazelcast-l337.ci.cloudbees.com/view/Official%20Builds/job/csharp-client-master/267/testReport/Hazelcast.Client.Test/ClientSSLTest/TestMapSSLEnabled_DoNotValidateName/

Hazelcast.Client.Test.ClientSSLTest.TestMapSSLEnabled_validateName_validName

MESSAGE:
                                                System.InvalidOperationException : Unable to connect to any address in the config! The following addresses were tried:127.0.0.1:5701
                                                +++++++++++++++++++
                                                STACK TRACE:
                                                at Hazelcast.Client.Spi.ClientClusterService.ConnectToOne() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Spi\ClientClusterService.cs:line 379
at Hazelcast.Client.Spi.ClientClusterService.ConnectToCluster() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Spi\ClientClusterService.cs:line 333
at Hazelcast.Client.Spi.ClientClusterService.Start() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Spi\ClientClusterService.cs:line 225
at Hazelcast.Client.HazelcastClient.Start() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client\HazelcastClient.cs:line 427
at Hazelcast.Client.HazelcastClient.NewHazelcastClient(ClientConfig config) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client\HazelcastClient.cs:line 329
at Hazelcast.Client.Test.ClientSSLTest.Setup(Boolean ssl, Boolean validateCertificateChain, Boolean validateCertificateName, String certificateName) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Test\Hazelcast.Client.Test\ClientSSLTest.cs:line 40
at Hazelcast.Client.Test.ClientSSLTest.TestMapSSLEnabled_validateName_validName() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Test\Hazelcast.Client.Test\ClientSSLTest.cs:line 54

https://hazelcast-l337.ci.cloudbees.com/view/Official%20Builds/job/csharp-client-master/267/testReport/Hazelcast.Client.Test/ClientSSLTest/TestMapSSLEnabled_validateName_validName/

[TEST-FAILURE] Hazelcast.Client.Test.ClientRetryTest.TestRetryAsyncRequest

https://hazelcast-l337.ci.cloudbees.com/job/csharp-client-pr-builder/270/

MESSAGE:
                                                Hazelcast.Core.HazelcastException : Object reference not set to an instance of an object.
                                                +++++++++++++++++++
                                                STACK TRACE:
                                                at Hazelcast.Client.Spi.ClientProxy.<>c__DisplayClass1`1.<InvokeAsync>b__0(Task`1 t) in c:\jenkins\ope-fs-root\workspace\csharp-client-pr-builder\Hazelcast.Net\Hazelcast.Client.Spi\ClientProxy.cs:line 77
at System.Threading.Tasks.ContinuationResultTaskFromResultTask`2.InnerInvoke()
at System.Threading.Tasks.Task.Execute()

Generic List other than List<Object> does not properly serialized

writing from c#

var map1 = client.GetMap<string, List<int>>("listTest");
map1.Set("vikTest", new List<int> {1, 2});

can’t read it from java

Exception in thread "main" com.hazelcast.nio.serialization.HazelcastSerializationException: There is no suitable de-serializer for type -110. This exception is likely to be caused by differences in the serialization configuration between members or between clients and members.

but it should use type -26 which is List

Listener registration/ re-registration is broken

An old bug in Java client fix should be ported in here too. The notion of failed-registration list is required.

Java to dotnet number serialization

Hello
I have tried to share some data between a java process and a dotnet one using a map.
Using a key of type double on both side of Integer/Int32 , I get serialization errors on the dotnet side with the following message "There is no suitable de-serializer for type 0".
I guess this might be related to dotnet "struct" types. I could wrap my key in a IPortable class but this does not seems to be a memory efficient solution...
Let me know if I have missed something in the doc
Regards

Listeners should listen on all nodes for better scalability

see: hazelcast/hazelcast#6475

Client does not respect long running operations

.Net client has a hard coded 250 sec of invocation timeout. This prevents submitting long running operations such as executeOnKey or complicated predicate queries.
After the timeout hit, waiting invocation responded by TimeoutException. But the server side continue to execute and respond the actual result. On this case;

Received an unhandled exception from a task!: System.InvalidOperationException: An attempt was made to transition a task to a final state when it had already completed. 
at System.Threading.Tasks.Task`1.set_Result(TResult value) 
at Hazelcast.Client.Spi.SettableFuture`1.set_Result(T value) 
at Hazelcast.Client.Spi.ClientInvocationService.HandleResponseMessage(IClientMessage response) 
at Hazelcast.Client.Spi.ClientInvocationService.<>c__DisplayClass4.<HandleClientMessage>b__1() 
at System.Threading.Tasks.Task.InnerInvoke() 
at System.Threading.Tasks.Task.Execute()

Java client has no similar timeout and waits the server side forever. The trivial solution is to implement the same behaviour.

InitialMembershipListener should not fire memberAdded events on startup

see: hazelcast/hazelcast#7543

.NET Standard 2.0 Support

CoreCLR cross-platform

Are there plans to support this? It would be awesome to use Hazelcast with Windows IoT for distributed computing. I've pulled down a Git copy of it, but it is a lot of stuff for me to learn and then convert over.

Simplify exception retryable checking by removing AuthenticationException from list.

hazelcast/hazelcast#7808

ClientCountDownLatchTest.TestLatch

                                                MESSAGE:
                                                Hazelcast.Core.HazelcastException : Hazelcast instance is not active!
TearDown : Hazelcast.Core.HazelcastException : Hazelcast instance is not active!
                                                +++++++++++++++++++
                                                STACK TRACE:
                                                at Hazelcast.Client.Spi.ClientProxy.Invoke(IClientMessage request, Object key) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Spi\ClientProxy.cs:line 163
at Hazelcast.Client.Proxy.ClientCountDownLatchProxy.Invoke(IClientMessage request) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Proxy\ClientCountDownLatchProxy.cs:line 61
at Hazelcast.Client.Proxy.ClientCountDownLatchProxy.TrySetCount(Int32 count) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Proxy\ClientCountDownLatchProxy.cs:line 55
at Hazelcast.Client.Test.ClientCountDownLatchTest.TestLatch() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Test\Hazelcast.Client.Test\ClientCountDownLatchTest.cs:line 46
--TearDown
at Hazelcast.Client.Spi.ClientProxy.Invoke(IClientMessage request, Object key) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Spi\ClientProxy.cs:line 163
at Hazelcast.Client.Proxy.ClientCountDownLatchProxy.Invoke(IClientMessage request) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Proxy\ClientCountDownLatchProxy.cs:line 61
at Hazelcast.Client.Spi.ClientProxy.Destroy() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Net\Hazelcast.Client.Spi\ClientProxy.cs:line 60
at Hazelcast.Client.Test.ClientCountDownLatchTest.Destroy() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Test\Hazelcast.Client.Test\ClientCountDownLatchTest.cs:line 35

https://hazelcast-l337.ci.cloudbees.com/view/Official%20Builds/job/csharp-client-master/172/testReport/junit/Hazelcast.Client.Test/ClientCountDownLatchTest/TestLatch/

Can not put byte array to IMap

When I do:
hzClient.getMap("map").Set("key", new byte[] {1,2,3});
It doest not work and my application hang on

Eventually Consistent Near Cache - Map

Map types listeners missing optimizations

Map type listeners are not inline with Java client implementation improvements.

• Events should be registered to sub listeners to not receive unnecessary events.
• Event objects are not lazy
• Event object cannot provide server merged value to client

ClientPartitionTest.TestPartitionsUpdatedAfterNewNode

                         MESSAGE:
                                                Could not verify assertion The partition list did not contain 2 partitions. after 60 seconds: NUnit.Framework.AssertionException:   Expected: 2
  But was:  1

   at NUnit.Framework.Assert.That(Object actual, IResolveConstraint expression, String message, Object[] args)
   at NUnit.Framework.Assert.AreEqual(Int32 expected, Int32 actual)
   at Hazelcast.Client.Test.HazelcastBaseTest.<AddNodeAndWait>b__2() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Test\Hazelcast.Client.Test\HazelcastBaseTest.cs:line 105
   at Hazelcast.Client.Test.TestSupport.AssertTrueEventually(Action asserAction, Int32 timeoutSeconds, String assertion) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Test\Hazelcast.Client.Test\TestSupport.cs:line 83
                                                +++++++++++++++++++
                                                STACK TRACE:
                                                at Hazelcast.Client.Test.TestSupport.AssertTrueEventually(Action asserAction, Int32 timeoutSeconds, String assertion) in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Test\Hazelcast.Client.Test\TestSupport.cs:line 92
at Hazelcast.Client.Test.HazelcastBaseTest.AddNodeAndWait() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Test\Hazelcast.Client.Test\HazelcastBaseTest.cs:line 108
at Hazelcast.Client.Test.ClientPartitionTest.TestPartitionsUpdatedAfterNewNode() in c:\jenkins\ope-fs-root\workspace\csharp-client-master\Hazelcast.Test\Hazelcast.Client.Test\ClientPartitionTest.cs:line 48

https://hazelcast-l337.ci.cloudbees.com/view/Official%20Builds/job/csharp-client-master/184/testReport/junit/Hazelcast.Client.Test/ClientPartitionTest/TestPartitionsUpdatedAfterNewNode/

Configuring System Properties Declaratively

ClientConfig should provide away for setting property and values

Use separate internal thread pool for handling messages

A separate internal thread pool should be used for handling incoming messages, rather than the default thread pool.

If the user creates a very large number of tasks (default thread pool size is 1024) which perform blocking operations on Hazelcast client, this can cause a deadlock as the Hazelcast operations also need a thread from the thread pool to process the response.

Exceptions while sending messages on reconnection

I have tried to test a failover scenario with hazelcast 3.6 and the csharp client (3.6 as well) and I think I might found a bug.
I have written a tiny csharp program that instantiate an hazelcast client that listens on a topic and send messages to the same topic within a loop. The client can connect to two different hazelcast instances.
The test scenario is quite simple :

1. start a first hazelcast instance
2. start a second hazelcast instance
3. start the csharp client
4. shutdown the hazelcast instance which have the longest uptime and then wait a few seconds
5. start again an hazelcast instance
6. wait a few seconds and iterate to step 4

After a few iterations my csharp client crashes while trying to send messages while switching from one instance to the other one.

I have also written the very same program in java - I have not succeed to reproduce the issue with the java version.

Below my csharp code:

 static void Main(string[] args)
    {
        Environment.SetEnvironmentVariable("hazelcast.logging.level", "All");
        Environment.SetEnvironmentVariable("hazelcast.logging.type", "console");
        var config = new ClientConfig();
        config.GetNetworkConfig().SetConnectionAttemptLimit(10);
        config.GetNetworkConfig().AddAddress(new[] { "127.0.0.1:5701", "127.0.0.1:5702" });
        var client = HazelcastClient.NewHazelcastClient(config);
        var topic = client.GetTopic<string>("backplane");
        topic.AddMessageListener(msg => Console.Out.WriteLine("Received msg: " + msg.GetMessageObject()));
        IAtomicLong counter = client.GetAtomicLong("counter");
        for (int i = 0; i < 1000; i++)
        {
            var sqn = counter.IncrementAndGet();
            Console.Out.WriteLine("Hello " + i + " " + sqn);
            Thread.Sleep(1000);
            topic.Publish(i + " hello " + DateTime.Now);
        }
     }

I start my hazelcast instances with the default config - i.e. with the following command line:

  start java -jar hazelcast-3.6.jar

Below an exception stacktrace I have got with the scenario described above:

    Hazelcast.Client.Spi.TargetDisconnectedException was unhandled
      _HResult=-2146233087
      _message=Target[Address[192.168.0.15]:5701] disconnected.
      HResult=-2146233087
      IsTransient=false
      Message=Target[Address[192.168.0.15]:5701] disconnected.
      Source=Hazelcast.Net
      StackTrace:
           à Hazelcast.Client.Spi.ClientProxy.Invoke(IClientMessage request, Object key)
           à Hazelcast.Client.Proxy.ClientTopicProxy`1.Invoke(IClientMessage request)
           à Hazelcast.Client.Proxy.ClientTopicProxy`1.Publish(T message)
           à HzSource.Program.Main(String[] args) dans c:\work\dotnetsandbox\HzSandbox\HzSource\Program.cs:ligne 41
           à System.AppDomain._nExecuteAssembly(RuntimeAssembly assembly, String[] args)
           à System.AppDomain.ExecuteAssembly(String assemblyFile, Evidence assemblySecurity, String[] args)
           à Microsoft.VisualStudio.HostingProcess.HostProc.RunUsersAssembly()
           à System.Threading.ThreadHelper.ThreadStart_Context(Object state)
           à System.Threading.ExecutionContext.RunInternal(ExecutionContext executionContext, ContextCallback callback, Object state, Boolean preserveSyncCtx)
           à System.Threading.ExecutionContext.Run(ExecutionContext executionContext, ContextCallback callback, Object state, Boolean preserveSyncCtx)
           à System.Threading.ExecutionContext.Run(ExecutionContext executionContext, ContextCallback callback, Object state)
           à System.Threading.ThreadHelper.ThreadStart()
      InnerException: 

Let me know if you need more details
Regards

Read/WriteObject in SerializationService non-compliant with Java equivalents

The ReadObject and WriteObject methods within the SerializationService class perform a null check on the object in question by either doing a ReadBoolean to determine if the object is null or a WriteBoolean to record the object as null.

ReadObject (C#)

try
            {
                var isNull = input.ReadBoolean();
                if (isNull)
                {
                    return default(T);
                }
....

WriteObject (C#)

var isNull = obj == null;
            try
            {
                output.WriteBoolean(isNull);
                if (isNull)
                {
                    return;
                }
                var serializer = SerializerFor(obj);
                output.WriteInt(serializer.GetTypeId());
                serializer.Write(output, obj);
            }
            catch (Exception e)
            {
                throw HandleException(e);
            }

The Java equivalents readObject and writeObject in the AbstractSerializationService class do not read or write any boolean from the stream to check for "nullness". The result of this is that any stream that utilizes these methods will fail because the C# client reads/writes one extra byte (for the "nullness" boolean). This causes reading and writing to fail. A quick check on the audit of both serialization service classes shows that the C# client has always done this and the Java equivalent hasn't done this, meaning these methods would have never worked. Updated: According to @cangencer this was changed in 3.5.5. I'm very new to Hazelcast (and C#/.NET itself) as well as Github (go easy on me :)), but can anyone confirm that my findings are indeed an issue? Removing these read/write boolean calls allows me to utilizes these methods.

Here are the Java equivalents to the above:

readObject

public final <T> T readObject(final ObjectDataInput in) {  
        try {  
            final int typeId = in.readInt();  
            final SerializerAdapter serializer = serializerFor(typeId);  
            if (serializer == null) {  
                if (active) {
                    throw newHazelcastSerializationException(typeId);
                }
                throw new HazelcastInstanceNotActiveException();
            }
            Object obj = serializer.read(in);
            if (managedContext != null) {
                obj = managedContext.initialize(obj);
            }
            return (T) obj;
        } catch (Throwable e) {
            throw handleException(e);
        }
    }

writeObject

@Override
    public final void writeObject(final ObjectDataOutput out, final Object obj) {
        if (obj instanceof Data) {
            throw new HazelcastSerializationException(
                    "Cannot write a Data instance! " + "Use #writeData(ObjectDataOutput out, Data data) instead.");
        }
        try {
            SerializerAdapter serializer = serializerFor(obj);
            out.writeInt(serializer.getTypeId());
            serializer.write(out, obj);
        } catch (Throwable e) {
            throw handleException(e);
        }
    }

Deserialization issue between Windows and Linux

There seems to be an issue using the .NET Client with Windows to Linux server, works with a Windows server. Could be a big-endian / little-endian difference.

Tech-stack:
.Net version 4.5
· Windows version: Windows 7 (for client) Windows server 2012 (if I am running Hazelcast server on Windpows)
· Hazelcast client version 3.7.1
· Linux version: Debian GNU/Linux8 (jessie)
· Hazelcast server version : 3.7.5
· Java :
o openjdk version "1.8.0_111"
o OpenJDK Runtime Environment (build 1.8.0_111-8u111-b14-2~bpo8+1-b14)
o OpenJDK 64-Bit Server VM (build 25.111-b14, mixed mode)

No further details yet...

Unhandled ThreadInterruptedException on I/O Writer thread

There exists an unhandled ThreadInterruptedException on I/O writer thread. Here is the stacktrace of the exception:

Exception Info: System.Threading.ThreadInterruptedException
Stack:
   at System.Threading.Monitor.ReliableEnter(System.Object, Boolean ByRef)
   at System.Threading.Monitor.Enter(System.Object, Boolean ByRef)
   at System.Resources.RuntimeResourceSet.GetObject(System.String, Boolean, Boolean)
   at System.Resources.RuntimeResourceSet.GetString(System.String, Boolean)
   at System.Resources.ResourceManager.GetString(System.String, System.Globalization.CultureInfo)
   at System.SR.GetString(System.String)
   at System.Collections.Concurrent.BlockingCollection`1[[System.__Canon, mscorlib, Version=4.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089]].Take()
   at Hazelcast.Client.Connection.ClientConnection.WriteQueueLoop()
   at System.Threading.ThreadHelper.ThreadStart_Context(System.Object)
   at System.Threading.ExecutionContext.Run(System.Threading.ExecutionContext, System.Threading.ContextCallback, System.Object, Boolean)
   at System.Threading.ExecutionContext.Run(System.Threading.ExecutionContext, System.Threading.ContextCallback, System.Object)
   at System.Threading.ThreadHelper.ThreadStart()

Map putAll/getAll implementation enhancement

Apply the changes in hazelcast/hazelcast#6555 and hazelcast/hazelcast#6480.

System.InvalidCastException when using PagingPredicate on Map.EntrySet()

When using a PagingPredicate with Map.EntrySet() (with returning some entries), the following exception is thrown at the client:

   at System.Linq.Enumerable.<CastIterator>d__94`1.MoveNext()
   at System.Collections.Generic.HashSet`1.UnionWith(IEnumerable`1 other)
   at System.Collections.Generic.HashSet`1..ctor(IEnumerable`1 collection, IEqualityComparer`1 comparer)
   at System.Collections.Generic.HashSet`1..ctor(IEnumerable`1 collection)
   at Hazelcast.Client.Proxy.ClientMapProxy`2.EntrySetWithPagingPredicate(PagingPredicate pagingPredicate)
   at Hazelcast.Client.Proxy.ClientMapProxy`2.EntrySet(IPredicate predicate)
   at ConsoleApplication3.Program.Main(String[] args) in c:\Users\Alparslan\Documents\Visual Studio 2013\Projects\ConsoleApplication3\ConsoleApplication3\Program.cs:line 32
   at System.AppDomain._nExecuteAssembly(RuntimeAssembly assembly, String[] args)
   at System.AppDomain.ExecuteAssembly(String assemblyFile, Evidence assemblySecurity, String[] args)
   at Microsoft.VisualStudio.HostingProcess.HostProc.RunUsersAssembly()
   at System.Threading.ThreadHelper.ThreadStart_Context(Object state)
   at System.Threading.ExecutionContext.RunInternal(ExecutionContext executionContext, ContextCallback callback, Object state, Boolean preserveSyncCtx)
   at System.Threading.ExecutionContext.Run(ExecutionContext executionContext, ContextCallback callback, Object state, Boolean preserveSyncCtx)
   at System.Threading.ExecutionContext.Run(ExecutionContext executionContext, ContextCallback callback, Object state)
   at System.Threading.ThreadHelper.ThreadStart()

The reproducer code is as follows:

var map = client.GetMap<int, int>("testMap");
var size = map.EntrySet(new PagingPredicate(3)).Count;

Map.KeySet() works smoothly. Client version is 3.8.2.

ReplicatedMap implementation

Leaking Handles when reconnecting

Handle leaking occur when reconnecting to server

can be reproduced by continually trying to connect to a nonexisting server

ClientConfig HazelConfig = null;
IHazelcastInstance instance = null;

        HazelConfig = new ClientConfig();
        HazelConfig.GetNetworkConfig().AddAddress("10.10.10.123");

        bool connected = false;
        while (!connected)
        {
            try
            {
                Program.instance = HazelcastClient.NewHazelcastClient(Program.HazelConfig);

                //                    ConnectToHazelcast();
                connected = true;
            }
            catch (Exception)
            {

            }
        }

Unhandled Exception: System.Threading.SynchronizationLockException

Unhandled Exception:

System.Threading.SynchronizationLockException: Object synchronization method was called from an unsynchronized block of code.
  at (wrapper managed-to-native) System.Threading.Monitor:Exit (object)
  at System.Threading.SemaphoreSlim.Wait (Int32 millisecondsTimeout, CancellationToken cancellationToken) [0x00179] in /private/tmp/source-mono-4.4.0-c7sr1/bockbuild-mono-4.4.0-branch-c7sr1/profiles/mono-mac-xamarin/build-root/mono-x86/external/referencesource/mscorlib/system/threading/SemaphoreSlim.cs:425 
  at System.Collections.Concurrent.BlockingCollection`1[T].TryTakeWithNoTimeValidation (System.Collections.Concurrent.T& item, Int32 millisecondsTimeout, CancellationToken cancellationToken, System.Threading.CancellationTokenSource combinedTokenSource) [0x00076] in /private/tmp/source-mono-4.4.0-c7sr1/bockbuild-mono-4.4.0-branch-c7sr1/profiles/mono-mac-xamarin/build-root/mono-x86/external/referencesource/System/sys/system/collections/concurrent/BlockingCollection.cs:712 
  at System.Collections.Concurrent.BlockingCollection`1[T].TryTake (System.Collections.Concurrent.T& item, Int32 millisecondsTimeout, CancellationToken cancellationToken) [0x00006] in /private/tmp/source-mono-4.4.0-c7sr1/bockbuild-mono-4.4.0-branch-c7sr1/profiles/mono-mac-xamarin/build-root/mono-x86/external/referencesource/System/sys/system/collections/concurrent/BlockingCollection.cs:667 
  at System.Collections.Concurrent.BlockingCollection`1[T].Take () [0x00000] in /private/tmp/source-mono-4.4.0-c7sr1/bockbuild-mono-4.4.0-branch-c7sr1/profiles/mono-mac-xamarin/build-root/mono-x86/external/referencesource/System/sys/system/collections/concurrent/BlockingCollection.cs:554 
  at Hazelcast.Client.Connection.ClientConnection.WriteQueueLoop () [0x0001e] in Hazelcast.Net/Hazelcast.Client.Connection/ClientConnection.cs:393 
  at System.Threading.ThreadHelper.ThreadStart_Context (System.Object state) [0x00017] in /private/tmp/source-mono-4.4.0-c7sr1/bockbuild-mono-4.4.0-branch-c7sr1/profiles/mono-mac-xamarin/build-root/mono-x86/external/referencesource/mscorlib/system/threading/thread.cs:68 
  at System.Threading.ExecutionContext.RunInternal (System.Threading.ExecutionContext executionContext, System.Threading.ContextCallback callback, System.Object state, Boolean preserveSyncCtx) [0x0008d] in /private/tmp/source-mono-4.4.0-c7sr1/bockbuild-mono-4.4.0-branch-c7sr1/profiles/mono-mac-xamarin/build-root/mono-x86/external/referencesource/mscorlib/system/threading/executioncontext.cs:957 
  at System.Threading.ExecutionContext.Run (System.Threading.ExecutionContext executionContext, System.Threading.ContextCallback callback, System.Object state, Boolean preserveSyncCtx) [0x00000] in /private/tmp/source-mono-4.4.0-c7sr1/bockbuild-mono-4.4.0-branch-c7sr1/profiles/mono-mac-xamarin/build-root/mono-x86/external/referencesource/mscorlib/system/threading/executioncontext.cs:904 
  at System.Threading.ExecutionContext.Run (System.Threading.ExecutionContext executionContext, System.Threading.ContextCallback callback, System.Object state) [0x00031] in /private/tmp/source-mono-4.4.0-c7sr1/bockbuild-mono-4.4.0-branch-c7sr1/profiles/mono-mac-xamarin/build-root/mono-x86/external/referencesource/mscorlib/system/threading/executioncontext.cs:893 
  at System.Threading.ThreadHelper.ThreadStart () [0x0000b] in /private/tmp/source-mono-4.4.0-c7sr1/bockbuild-mono-4.4.0-branch-c7sr1/profiles/mono-mac-xamarin/build-root/mono-x86/external/referencesource/mscorlib/system/threading/thread.cs:105 

test

Library does not use Properties and instead uses Getter Functions

The lack of Properties breaks a huge number of features in the .net framework.

ClientTxnMapTest Failures

TestKeySetValues

MESSAGE:
                                                  Expected: 3
  But was:  2

                                                +++++++++++++++++++
                                                STACK TRACE:
                                                at Hazelcast.Client.Test.ClientTxnMapTest.TestKeySetValues() in c:\jenkins\ope-fs-root\workspace\csharp-client-maintenance\Hazelcast.Test\Hazelcast.Client.Test\ClientTxnMapTest.cs:line 130

TestPutGet

MESSAGE:
                                                System.InvalidOperationException : Nested transactions are not allowed!
                                                +++++++++++++++++++
                                                STACK TRACE:
                                                at Hazelcast.Util.ExceptionUtil.Rethrow(Exception t) in c:\jenkins\ope-fs-root\workspace\csharp-client-maintenance\Hazelcast.Net\Hazelcast.Util\ExceptionUtil.cs:line 118
at Hazelcast.Client.Request.Transaction.TransactionProxy.Begin() in c:\jenkins\ope-fs-root\workspace\csharp-client-maintenance\Hazelcast.Net\Hazelcast.Client.Request.Transaction\TransactionProxy.cs:line 72
at Hazelcast.Client.Request.Transaction.TransactionContextProxy.BeginTransaction() in c:\jenkins\ope-fs-root\workspace\csharp-client-maintenance\Hazelcast.Net\Hazelcast.Client.Request.Transaction\TransactionContextProxy.cs:line 43
at Hazelcast.Client.Test.ClientTxnMapTest.TestPutGet() in c:\jenkins\ope-fs-root\workspace\csharp-client-maintenance\Hazelcast.Test\Hazelcast.Client.Test\ClientTxnMapTest.cs:line 40

https://hazelcast-l337.ci.cloudbees.com/view/Official%20Builds/job/csharp-client-maintenance/560/

Prevent buffer resizing when reading ClientMessages

hazelcast/hazelcast#7547 and hazelcast/hazelcast#7550

Logging improvements on .NET Client

This is what we do on the server. We need to provide the same level of version information on the clients.

INFO: [10.0.0.2]:5701 [dev] [3.6-EA2] Hazelcast 3.6-EA2 (20151125 - 8af85c5) starting at Address[10.0.0.2]:5701

• 3.6-EA2 is the version.
• Hazelcast 3.6-EA2 is the full name
• 20151125 is the release date
• 8af85c5 is the GitHub Id.

See https://github.com/hazelcast/hazelcast/releases. You will see it corresponds to EA2

Hang for ever with Hazelcast 3.5

Hello,
Thanks for having open sourced this csharp client.
I have done some testing and I have come to the conclusion that the current csharp client can only connect to an hazelcast cluster using a 3.6-RC snapshot version. Otherwise using Hazelcast 3.5, the client hangs for ever.
Am I missing something?
Regards