by

Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • Jul 05 18:52
    KarolKalbarczyk commented #3234
  • Jul 03 20:21
    Arkatufus edited #4500
  • Jul 03 20:20
    Arkatufus opened #4500
  • Jul 03 18:12
    ismaelhamed opened #4499
  • Jul 03 17:40
    ismaelhamed opened #4498
  • Jul 02 19:33
    Danthar commented #4487
  • Jul 02 19:27
    Danthar closed #4494
  • Jul 02 19:27
    Danthar commented #4494
  • Jul 02 16:12
    Arkatufus commented #4265
  • Jul 02 15:18
    ismaelhamed edited #4497
  • Jul 02 15:16
    ismaelhamed edited #4497
  • Jul 02 15:15
    ismaelhamed opened #4497
  • Jul 02 15:07
    ismaelhamed commented #4482
  • Jul 02 08:02
    glikoz commented #4495
  • Jul 02 08:02
    glikoz commented #4495
  • Jul 01 13:26
    Aaronontheweb closed #4495
  • Jul 01 13:26
    Aaronontheweb commented #4495
  • Jul 01 13:20
    Aaronontheweb commented #4496
  • Jul 01 13:20
    Aaronontheweb milestoned #4496
  • Jul 01 13:20
    Aaronontheweb labeled #4496
Stijn Herreman
@stijnherreman
@Aaronontheweb thanks! I was stuck on this for a couple of hours already, trying a bunch of different things.
Aaron Stannard
@Aaronontheweb
did that suggestion work?
Stijn Herreman
@stijnherreman
Yes, see the comment to your answer
Aaron Stannard
@Aaronontheweb
ah nice
good deal, I'd been getting behind in racking up my SO karma
Stijn Herreman
@stijnherreman
Heh :) on that topic, not that I have a shortage of rep but feel free to upvote the questions :)
Andrew Young
@ayoung
@Danthar thanks. what about with the default wire format or hyperion?
Arsene
@Tochemey
Hello can someone educate on NBench and probably what is its use and a tutorial.
Natan Vivo
@nvivo
What is the best way to detect if the node is unreachable from within the unreachable node? Will it see the other nodes as unreachable or does it set any status internally to detect this?
I was actually wondering is there is a way to completely avoid the idea of unreachable nodes, and make it just reconnect. I know currently the cluster algorithm depends on that but I'm coming to the conclusion it's simpler to not have these controls and deal with the other issues than deal with unreachable nodes
The fact is that the cloud is a very unstable place, and as the number of nodes go up, I'm having unreachable nodes twice a day now due to conditions I cannot control. These nodes eventually come back, but while that is happening the cluster gets unusable.
Natan Vivo
@nvivo
Created a discussion on #3144
would like to hear some thoughts on that
Bartosz Sypytkowski
@Horusiath
@nvivo if you think of cluster as a graph, unreachability is a property of a graph's edge, not a node itself (nor a subset of the graph)
in general if A sees B as unreachable, B should see A as unreachable too, but this may not happen at exactly the same moment
also AFAIK unreachable nodes are trying to reconnect to each other every 5s (by default)
also #3099 should help a little - it introduces an extra node status, that will allow you to use nodes even before cluster reaches full convergence (which is hard to achieve when node churn is big)
carnogursky
@carnogursky
Hi, please, could someone advise how to implement SSL certificates to the TCP communication (Akka.IO)?
Bartosz Sypytkowski
@Horusiath
@carnogursky I'm working on it. The closest approach would be to implement DotNetty's style TlsHandler as akka actor
carnogursky
@carnogursky
@Horusiath Thank you. Please, could you give me some more detailed hint? I do not know how TslHandler in DotNetty works...
Bartosz Sypytkowski
@Horusiath
@carnogursky this is a pretty complex stuff, I'm not sure if you want to dig deep into it. TLS needs its handshaking and secure stream handled - so far, only mature API for SSL in .NET is SslStream. However in akka actors and IO streams talks via ByteString. Because of that we need an adapter stream between actor and SslStream.
carnogursky
@carnogursky
@Horusiath It looks like DotNetty has a solution, which can be implemented easier? Had checked theis examples of server, you suggested it coulf be implemented as Actor?
Bartosz Sypytkowski
@Horusiath
DotNetty has a solution, but it's easier only if you want to use DotNetty directly
Jalal EL-SHAER
@jalchr
@Horusiath in DistributedData ... why would a 'Get' operation throw exception rather than returning null for a specific key ?
2017-10-09 01:17:37,203 [71] ERROR Akka.Actor.OneForOneStrategy - One or more errors occurred.
System.AggregateException: One or more errors occurred. ---> Akka.DistributedData.DataDeletedException: Cannot retrieve data under key [146324.mpg]. It has been permanently deleted and the key cannot be reused.
   at Akka.DistributedData.DistributedData.<GetAsync>d__13`1.MoveNext()
   --- End of inner exception stack trace ---
   at System.Threading.Tasks.Task.ThrowIfExceptional(Boolean includeTaskCanceledExceptions)
   at System.Threading.Tasks.Task`1.GetResultCore(Boolean waitCompletionNotification)
   at System.Threading.Tasks.Task`1.get_Result()
   at Archive.VideoProcessing.Akka.VideoProviderWatcherActor.<Watch>b__10_4(FileFinished msg)
   at lambda_method(Closure , Object , Action`1 , Action`1 , Action`1 , Action`1 , Action`1 , Action`1 , Action`1 , Action`1 , Action`1 , Action`1 )
   at Akka.Tools.MatchHandler.PartialHandlerArgumentsCapture`11.Handle(T value)
   at Akka.Actor.ReceiveActor.ExecutePartialMessageHandler(Object message, PartialAction`1 partialAction)
   at Akka.Actor.ReceiveActor.OnReceive(Object message)
   at Akka.Actor.UntypedActor.Receive(Object message)
   at Akka.Actor.ActorBase.AroundReceive(Receive receive, Object message)
   at Akka.Actor.ActorCell.ReceiveMessage(Object message)
   at Akka.Actor.ActorCell.Invoke(Envelope envelope)
---> (Inner Exception #0) Akka.DistributedData.DataDeletedException: Cannot retrieve data under key [146324.mpg]. It has been permanently deleted and the key cannot be reused.
   at Akka.DistributedData.DistributedData.<GetAsync>d__13`1.MoveNext()<---
Stijn Herreman
@stijnherreman
@alexvaluyskiy any idea if it's possible to have packages that can target netstandard1.6 (like Akka.Remote), additionally also target netstandard2.0? The NETStandard.Library 1.6 package is a meta package and suffers from dependency bloat (few dozen dependencies), this was fixed in the 2.0 package. Since my application targets net461, I managed to work around this by first manually installing NETStandard.Library 2.0 and then updating Akka.Remote.
But probably not every users realises they can do this, so perhaps it's possible to help them out. I'm not sure though if NuGet is smart enough to pick the highest possible target, when installing a package (e.g. it should pick netstandard2.0 if possible, over netstandard1.6)
Bartosz Sypytkowski
@Horusiath
@jalchr
  1. null should be returned in v1.3.1
  2. DistributedData methods are meant to be used outside actor context, inside actors using traditional message passing is a lot faster. Also message based API is bigger and gives more power i.e. you can subscribe to receive updates of target CRDT as they come.
  3. Keep in mind that DistributedData is not meant for high volume of data. If you want to use it for message delivery, you may eventually run out of memory.
Jalal EL-SHAER
@jalchr
@Horusiath I'm using DistributedData to track state across multiple nodes that share same 'file processing'. I'm using it like a "Shared Dictionary". So its a kind of acknowledgement at the business level. The processing of files takes time, so performance is not an issue.
When you say "run-out-of-memory", do you mean that state is never cleaned ? I'm ensuring that I remove everything that is marked as "done processing". Are we safe ?
OysteinKoppang
@OysteinKoppang

Hi. I'm trying to secure a akka.net remote system with the new SSL support.

There are two actor systems, that both talk to each other with remote actor selections.

System A's config:

remote {
  dot-netty.tcp {
    port = 11293
    hostname = "127.0.0.1"
    enable-ssl = true
    maximum-frame-size = 4000000b
    ssl {
      suppress-validation = true
        certificate {
          path = "C:\\foo\\bar.pfx" 
          password = "baz"
          flags = ["user-key-set"]
        }
    }
  }
}

System B's config:

"remote": {
  "dot-netty.tcp": {
    "port": 11292,
    "hostname": "127.0.0.1",
    "enable-ssl": true,
    "maximum-frame-size": "4000000b",
    "ssl": {
      "suppress-validation": true,
      "certificate": {
        "path": "C:\\foo\\bar.pfx",
        "password": "baz",
        "flags": [ "user-key-set" ]
      }
    }
  }
}

When system B tries to do an actor selection using the address:
akka.ssl.tcp://FooSystem@127.0.0.1:11293/user/FooActor

An error is thrown:
'No transport is loaded for protocol: [akka.ssl.tcp], available protocols: [akka.tcp]'

Any ideas why this is happening?

Aaron Stannard
@Aaronontheweb
weird
no idea offhand
would you please file an issue? We'll try to reproduce it
Arturo Sevilla
@arturosevilla
Hi, I've introduced a behavioral change in the Serilog adapter, which I believe is currently inconsistent with the docs, and now is more ad-hoc with how you expect a logger to behave, could someone care to comment on this change? akkadotnet/Akka.Logger.Serilog#33
Deniz İrgin
@Blind-Striker

hi, we updated our applications akka version from 1.1.3 to 1.3.1 and change helios to dot-netty.tcp. We also updated Akka.Serialization.Hyperion from 1.1.3.32-beta to 1.3.1-beta.

After we started our cluster we getting error below ;

Cannot find serializer with id [9]. The most probable reason is that the configuration entry 'akka.actor.serializers' is not in sync between the two systems.

        serializers {
          hyperion = "Akka.Serialization.HyperionSerializer, Akka.Serialization.Hyperion"
        }

        serialization-bindings {
         "System.Object" = hyperion
        }
Bartosz Sypytkowski
@Horusiath
@Blind-Striker serializerId=9 is by default used by distributed pub sub message serializer. Check if you for sure added fallback to DistributedPubSub.DefaultConfig()
Deniz İrgin
@Blind-Striker
@Horusiath Where should we add this configuration. We using hocon for configuring our applications
Bartosz Sypytkowski
@Horusiath
you can put it as a fallback right before sending configuration to actor system
murat mert
@muratmert
@Blind-Striker can you add actor config section ?
Deniz İrgin
@Blind-Striker
So do i need to add this lines before creating actor system ;
  var section = (AkkaConfigurationSection)ConfigurationManager.GetSection("akka");
        var clusterConfig = section.AkkaConfig;
        clusterConfig.WithFallback(DistributedPubSub.DefaultConfig());

        _actorSystem = ActorSystem.Create("notificationSystem", clusterConfig);
@muratmert i have 3 different nodes
Deniz İrgin
@Blind-Striker
@Horusiath we added DistributedPubSub.DefaultConfig() as fallback but we still getting same error
Stijn Herreman
@stijnherreman
I have an actor that needs to handle one type of message only, RetrieveFoo, to retrieve a resource from a REST API. Once the resource is retrieved, it is cached indefinitely. Is this a good case for using async/await (and thus blocking the actor), or should I still attempt to use PipeTo and keep track in some way of RetrieveFoo messages that need to be responded to?
I want to avoid multiple requests to retrieve the resource.
Stijn Herreman
@stijnherreman
After thinking about this some more, I think I can just store a task in a variable when receiving the first message and use PipeTo, and then for all subsequent messages skip creating the task and just use PipeTo on the existing task. That should work even when the task is completed already.
Stijn Herreman
@stijnherreman
I came up with the following (initial) implementation. It feels a bit awkward to use ContinueWith but maybe that's of my habit to use async/await.
    public sealed class Specification
        : ReceiveActor
    {
        private readonly IClient specsDataServiceClient;

        private Task<Messages.Models.Specification> modelTask;

        public Specification(IClient specsDataServiceClient)
        {
            this.specsDataServiceClient = specsDataServiceClient ?? throw new ArgumentNullException(nameof(specsDataServiceClient));

            this.Receive<GetSpecification>(message => this.GetSpecification(message));
        }

        private void GetSpecification(GetSpecification message)
        {
            if (modelTask == null)
            {
                var getSpecificationTask = this.specsDataServiceClient.Specification_GetSpecificationAsync(message.SpecificationId);
                var getOperationsTask = this.specsDataServiceClient.Operations_GetOperationsForSpecIdAsync(message.SpecificationId);
                var getTestsTask = this.specsDataServiceClient.Test_GetTestsForSpecIdAsync(message.SpecificationId);

                var modelTask = Task.WhenAll(getSpecificationTask, getOperationsTask, getTestsTask)
                    .ContinueWith(_ =>
                    {
                        return new Messages.Models.Specification(getSpecificationTask.Result, getOperationsTask.Result, getTestsTask.Result);
                    });
            }

            var senderClosure = this.Sender;
            modelTask.PipeTo(senderClosure);
        }
    }
Arsene
@Tochemey
@stijnherreman Since you are using Actor if it is possible simply use the synchronous version of your function call.
@stijnherreman Also if you can explain in details what you want to do I can be of a help.
Stijn Herreman
@stijnherreman
@Tochemey no sync versions available unfortunately. The IClient implementation code is generated by a third-party tool, from an OpenAPI (Swagger) spec.
Arsene
@Tochemey
@stijnherreman Okay