Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • Oct 21 15:57
    Aaronontheweb closed #3877
  • Oct 21 15:57
    Aaronontheweb commented #3877
  • Oct 21 15:56

    Aaronontheweb on dev

    Persistence TestKit documentati… (compare)

  • Oct 21 15:56
    Aaronontheweb closed #3889
  • Oct 21 07:27
    dependabot-preview[bot] labeled #3999
  • Oct 21 07:27

    dependabot-preview[bot] on nuget

    Bump FluentAssertions from 4.14… (compare)

  • Oct 21 07:27
    dependabot-preview[bot] opened #3999
  • 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
Bartosz Sypytkowski
@Horusiath
ok, I'm merging that
Aaron Stannard
@Aaronontheweb
just about have the PostgreSql test suite working on Mono
have to do a bunch of funky Linux-specific file globbing and crap
ty @Horusiath
Aaron Stannard
@Aaronontheweb
alright, I give up - between half the shit for Akka.Persistence.PostgreSql not working at all on Mono and Docker on Windows Server 2016 barfing up segfaults when I try to run postgres on it
I've exhausted the options for getting builds running for it again
right way to fix this is to force all Akka.Persistence plugins to work on Mono only for the time being (i.e. build chain, drivers work on Mono or no CI) - SQL Server, on the other hand, we should be able to get working on 2016
Aaron Stannard
@Aaronontheweb
alright, Mono build chain is able to run end to end except for some Akka.IO specs
more TCP stuff again
going to see if I can get some more information with the logging settings turned up
if that doesn't help me suss out what the issue is exactly, going to mark it as "won't fix" since all of this shit is about to be rewritten anyway
Aaron Stannard
@Aaronontheweb
woot
think I was able to hotfix Akka.IO
riawz
@riawz

does it make any difference in terms of resource sharing / performance in an actor which does self polling

to have a short delay using Context.System.Scheduler.ScheduleTellOnce(10, Self, SelfPoll(), Self), like 10 ms
VS
Self.Tell(SelfPoll()) straight away

Aaron Stannard
@Aaronontheweb
@riawz the IScheduler allocates a Task-like object basically
the SelfPoll will fire immediately without allocating anything other than the message (and possibly some space inside the mailbox depending on how large the queue is)
really though, I'd use the scheduler if you really need a delay of any kind
otherwise it's safe to just to a self tell right away
@Silv3rcircl3 got the Akka.Streams suite mostly passing on Mono now... had a weird compilation bug on one spec that broke on a rebase
two areas where Reflection.Emit (from inside the expression compiler) power vomits and blows things up
going to see if there's a work-around
riawz
@riawz
@Aaronontheweb thanks!
Aaron Stannard
@Aaronontheweb
@Silv3rcircl3 anything special about these specs that failed in Akka.Streams? Trying to find the common root cause given that the failure is the same, but there are other specs inside these test fixtures which pass: http://petabridge-ci.cloudapp.net/viewLog.html?buildId=15933&tab=buildResultsDiv&buildTypeId=AkkaNet_LinuxMonoBuild&guest=1
this is building off of a personal branch of mine at the moment - haven't fundamentally changed streams in any meaningful way
Boban
@bobanco
This message was deleted
Aaron Stannard
@Aaronontheweb
what it looks like to me
given that the error messages are all the same
System.InvalidOperationException : 
  at (wrapper managed-to-native) System.Reflection.Emit.DynamicMethod:create_dynamic_method (System.Reflection.Emit.DynamicMethod,System.Reflection.Emit.DynamicMethod)
  at System.Reflection.Emit.DynamicMethod.CreateDynMethod () <0x404d7590 + 0x0019e> in <filename unknown>:0 
  at System.Reflection.Emit.DynamicMethod.CreateDelegate (System.Type delegateType, System.Object target) <0x404d74f0 + 0x0003a> in <filename unknown>:0 
  at System.Linq.Expressions.Compiler.LambdaCompiler.CreateDelegate () <0x404d7370 + 0x0008a> in <filename unknown>:0 
  at System.Linq.Expressions.Compiler.LambdaCompiler.Compile (System.Linq.Expressions.LambdaExpression lambda, System.Runtime.CompilerServices.DebugInfoGenerator debugInfoGenerator) <0x404c5730 + 0x000bb> in <filename unknown>:0 
  at System.Linq.Expressions.Expression`1[TDelegate].Compile () <0x404c56e0 + 0x0000f> in <filename unknown>:0 
  at Akka.Util.Reflection.ExpressionExtensions.GetArguments (System.Linq.Expressions.NewExpression newExpression) <0x404c38b0 + 0x0011f> in <filename unknown>:0 
  at Akka.Actor.Props.Create[TActor] (System.Linq.Expressions.Expression`1 factory, Akka.Actor.SupervisorStrategy supervisorStrategy) <0x404c3690 + 0x000d7> in <filename unknown>:0 
  at Akka.Actor.LocalActorRefProvider+<CreateSystemGuardian>c__AnonStorey1.<>m__0 () <0x404bfe00 + 0x00143> in <filename unknown>:0 
  at Akka.Actor.LocalActorRefProvider.CreateRootGuardianChild (Akka.Actor.LocalActorRef rootGuardian, System.String name, System.Func`1 childCreator) <0x404b4970 + 0x0003e> in <filename unknown>:0 
  at Akka.Actor.LocalActorRefProvider.CreateSystemGuardian (Akka.Actor.LocalActorRef rootGuardian, System.String name, Akka.Actor.LocalActorRef userGuardian) <0x404bf5b0 + 0x001f3> in <filename unknown>:0 
  at Akka.Actor.LocalActorRefProvider.Init (Akka.Actor.Internal.ActorSystemImpl system) <0x404b02a0 + 0x00297> in <filename unknown>:0 
  at Akka.Actor.Internal.ActorSystemImpl.Start () <0x404afb50 + 0x000e8> in <filename unknown>:0
there's some sort of actor we're using in the test fixture for these Akka.Streams tests
which might have something that is lazily initialized or asynchronously started
and there's a race condition by which if something else happens during actor system startup before that lazy loading completes, some part of the actor creation process for the built-in actors in the fixture itself fails
are there any special test actors that these fixtures use by default in the Akka.Streams.Tests library?
going to test this theory by running the suite again
and seeing if the spec failures are any different
any ideas @Silv3rcircl3 ?
Marc Piechura
@marcpiechura
@Aaronontheweb not out of my head, haven't seen this error ever before
I need to look into the specs but as far as I remember these "one delayed one failed publisher" specs are all in a base class and you need to implement a factory method which is called from the tests
So you probably only need to fix it in one place ;)
I don't have time the next two days but I could look into it over the weekend if you haven't fixed it until then
Peter Hvidgaard
@hvidgaard
If I have an actor that must update the UI, must I create it in HOCON or can I use it programmatically without config?
Arjen Smits
@Danthar
You can pass the correct dispatcher programmatically
Peter Hvidgaard
@hvidgaard
@Danthar

If I use

Context.ActorOf(Context.DI().Props<UiWorkerActor>().WithDispatcher("akka.actor.synchronized-dispatcher"), "ui-worker");

I get the following error: "The current SynchronizationContext may not be used as a TaskScheduler."

Am I missing something?
I'm using 1.1.1.27
Bartosz Sypytkowski
@Horusiath
@hvidgaard you're probably using async somewhere in your UiWorkerActor
you cannot combine that with synchronized-dispatcher
Peter Hvidgaard
@hvidgaard
@Horusiath The error comes when the child actor is created.
Bartosz Sypytkowski
@Horusiath
I would guess, that one of the actor or its children is using something like async version of Receive
Peter Hvidgaard
@hvidgaard
It only calls Recieve<>(), no async code.