Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • Oct 20 17:25
    valdisz synchronize #3889
  • Oct 20 17:17
    valdisz synchronize #3889
  • Oct 20 15:44
    valdisz synchronize #3889
  • Oct 20 09:22
    ismaelhamed commented #3863
  • Oct 19 23:39
    valdisz synchronize #3889
  • Oct 19 23:08
    edvinasz commented #2947
  • Oct 19 13:36
    Aaronontheweb commented #3973
  • Oct 19 13:34
    dependabot-preview[bot] synchronize #3995
  • Oct 19 13:34

    dependabot-preview[bot] on nuget

    Bump BenchmarkDotNet from 0.10.… (compare)

  • Oct 19 13:34
    dependabot-preview[bot] edited #3995
  • Oct 19 13:34
    dependabot-preview[bot] synchronize #3993
  • Oct 19 13:34

    dependabot-preview[bot] on nuget

    Bump Google.Protobuf from 3.9.1… (compare)

  • Oct 19 13:34
    dependabot-preview[bot] synchronize #3991
  • Oct 19 13:34

    dependabot-preview[bot] on nuget

    Bump Microsoft.Extensions.Depen… (compare)

  • Oct 19 13:34
    dependabot-preview[bot] synchronize #3989
  • Oct 19 13:34

    dependabot-preview[bot] on nuget

    Bump ApiApprover from 3.0.1 to … (compare)

  • Oct 19 13:34
    dependabot-preview[bot] synchronize #3992
  • Oct 19 13:34
    dependabot-preview[bot] edited #3993
  • Oct 19 13:34
    dependabot-preview[bot] synchronize #3985
  • Oct 19 13:34

    dependabot-preview[bot] on nuget

    Bump System.Reflection.Emit fro… (compare)

Bartosz Sypytkowski
@Horusiath
if your actor system responds with expected result on given request, then how it's implemented is a secondary matter
this way, you can still reimplement it without need to changing the tests
Thomas Tomanek
@thomastomanek
@Horusiath yep totally agree with you in general, but in this case the actor doesn't really have any output in terms of messages but you'd still want to verify it's doing what you want (given it's importance). I agree it's best to avoid peering into any classes internal state to validate behavior, but then again it's a rule I break on occasion when necessary
Thomas Tomanek
@thomastomanek
maybe I've missed a concept and have this one master actor doing too much and should get it to send out messages to create the domain objects
Roger Johansson
@rogeralsing
You can do state testing using .TestActorOf then you have access to the ubderlaying actor and its public members
So you could potentially check if the actor has a new child
Using testkit that is
Thomas Tomanek
@thomastomanek
@rogeralsing ah cool I'll take a look thanks
Roger Johansson
@rogeralsing
Using TestActorOf is a little bit different than what we are used to with actors, it will apply a synchronous dispatcher so all Tells are sync/blocking so that you can do state observations... so there are essentially two flavors of testabillity in TestKit, either with messages and Expect* or sync state testing using TestActorOf
Roger Johansson
@rogeralsing
ActorOfAsTestActorRef is the correct method name
in TestKitBase
NeoDarque
@NeoDarque
I’m playing around with Akka.Cluster and have managed to get a cluster router up and running which I can use to send messages to the relevant nodes when they join the cluster. However, I’ve got the configuration of the router working in code only and would like to move it to the hocon-config file. How would I translate the following code to config?
system.ActorOf(Props.Empty.WithRouter(
    new ClusterRouterGroup(
        new RoundRobinGroup("/user/backend"),
        new ClusterRouterGroupSettings(10, false, "backend", ImmutableHashSet.Create("/user/backend"))
    )
));
Weston
@ronnyek
@rogeralsing - interesting comparison... I was trying to build a sort of distributed pool of workers, with the management and error kernel being a bonus. Sounds like orleans is specifically for minimal impact state management of as many grains as possible... not a worker that might awake for a workload and spend a couple minutes working on it, and then perform some notification on completion
Bartosz Sypytkowski
@Horusiath
@ronnyek what is your case specifically?
Weston
@ronnyek
I've effectively got data coming in from different sources, sometimes polled, sometimes pulled, sometimes pushed in.
Ultimately I need to see what of that data a customer is looking for and notify them in the appropriate manner
so it woudlnt be like oh we processed this data, now send notification to 20k users at the same time
it'd be much more specific
I figure if I could have a pool of processors, and a pool of notifiers
because notifiers need to determine best method to notify client, and actually do the notification
I oculd just have a big mq, and a service hosted that just consumes work items off the queue
and notify accordingly, but seems like if the lifecycle of actors in akka made sense, this owuld be much better for being able to scale out resources
we're having a lot more processing consumption, increase resources there
Weston
@ronnyek
am I way off base? I know OF projects that use akka / akka.net, but I dont know what responsibilities akka serves there... or how it helps those projects scale.
also, are akka/akka.net really a major version or more behind, or do they just have diff versioning?
Timur Babyuk
@timba
Aaron Stannard
@Aaronontheweb
found a regression in Akka.Remote we introduced in v1.0.2
serialization problem with actor refs. Think it has to do with NoSender going over the network but I'm not 100% sure yet.
@ronnyek we have totally different versioning than they do
we started working on Akka.NET somewhere around Akka 2.1-2.2
Roger Johansson
@rogeralsing
@Aaronontheweb how does the regression manifest itself?
Roger Johansson
@rogeralsing
@Aaronontheweb btw. "RemoteDeploySpec" is wrong, it doesnt do anything remote. everything is just deployed locally
or am I missing the purpose of it ?
Roko Bobic
@rocco337
Hi, I need help with scheduller
Can I cancell scheduler from actor?
Boban
@bobanco
@rocco337 Yes you can cancel the scheduler, but you have to use Context.System.Scheduler.Advanced.ScheduleRepeatedlyCancelable which will return ICancelable and then you can call .Cancel() on that.
Bartosz Sypytkowski
@Horusiath
@rocco337 when you subscribe to scheduler you can pass ICancellable to it or receive cancellable object as a subscribe result
Roger Johansson
@rogeralsing
ScheduleTellRepeatedlyCancelable and ScheduleTellOnceCancelable are available directly on the standard scheduler
Roko Bobic
@rocco337
Tnx. I understand that you can cancel scheduler. But can you cancel scheduler from actor that is scheduled?
Basically I triggered repeated scheduler to send message on one actor
and than fro mthat actor I would like to cancel scheduler
Roger Johansson
@rogeralsing
oh, no, not unless you pass the ICancellable to it
Roko Bobic
@rocco337
aha yes I thought so. Maybe than it is not good idea to try to cancel itfro mthere :)
Roko Bobic
@rocco337
@rogeralsing just read your blog about serviceBus. It is jus what I needed.Tnx
Bartosz Sypytkowski
@Horusiath
ok @rogeralsing . I think I've covered #1067
Aaron Stannard
@Aaronontheweb
@rogeralsing an all-systems address confused the RemoteActorRefProvider and it tried to resolve a local address for an akka:// actor path
which is by definition, local
the remote deploy spec is testing the deployer, not the actual over-the-wire deployment mechanism
there's multi-node tests for Akka.Remote and the RemoteRouterSpec which actually do real remote deployments