dotnet/orleans

Orleans - Distributed Actor Model

Sergey Bykov

@sergeybykov

@Zeroshi

hi @sergeybykov , I was checking in to see if you had a chance to read that essay.

Sorry, I dropped the ball when we suddenly got some snow. I read it now. I think it'd be better for us to chat thematically over Teams/Skype than for me to try to type the same thoughts, and then for us to go back and forth. A voice dialog is an efficient means of arriving to something. 🙂

COCPORN

@COCPORN

@sergeybykov Yes, it was hyperbole from me. About the buckets, perhaps I've misunderstood how they work. It looks to just be sorted on expiry time, and then I suppose any grain that has activity moves itself to another bucket based on the time to live configuration. I could take a stab at implementing my suggestion, as I think the amount of work needed to have something that is useful (or can at least be evaluated for usefulness) isn't that big.

Veikko Eeva

@veikkoeeva

This https://github.com/davidfowl/Micronetes looks like a way to do https://github.com/dotnet/orleans/tree/master/Samples/OneBoxDeployment like things in the future.

Veikko Eeva

@veikkoeeva

Interesting, interesting... https://blog.acolyer.org/2020/01/20/stella-coping-with-complexity-2/

Panu Oksala

@PanuOksala_twitter

I'm running heterogeneous cluster with multiple Silos and it seems to be that all the Silos need to reference entities that are exposed in grain interfaces, is this true?

basically I need to have shared entity library that is referenced from all the Silos. This breaks the microservice architecture

If I don't reference this shared library from one the Silos, the grain calls may crash in binary serialization, because the passing through silo don't know about entities

COCPORN

@COCPORN

Perhaps I am misunderstanding, I am not sure what you mean by heterogeneous cluster if the silos are different. How do you expect the silo to be able to deserialize something that it doesn't have an implementation class for it?

Panu Oksala

@PanuOksala_twitter

I except that when calling Silo that does not have implementation for the actor, it would pass it to another Silo which contains the implementation for actor, without deserializing the method parameters

Veikko Eeva

@veikkoeeva

Heh, mentioned Orleans indirectly at https://twitter.com/hiddenmarkov/status/1219288129716334592 and https://twitter.com/_dotnetbot_ picked it up. :D Though the truth is, I believe, Orleans and actor model in general, could fit even complex CPS deployments. There just are gnarly problems otherwise to consider.

Panu Oksala

@PanuOksala_twitter

Silo A hosts Grain A1
Silo B hosts Grain B1 which has method DoSomething(Customer customer)

I call grain B1 from clusterclient, the invocation goes to A1 which deserializes the message (and crashes because it does not know about Customer) and does not resend the message to B1

COCPORN

@COCPORN

@PanuOksala_twitter I don't know about this kind of behavior. Is it a documented feature?

Panu Oksala

@PanuOksala_twitter

If I add Customer class into shared dll and add reference to Silo A, everything works.

Documentation says: All silos should reference interfaces of all grain types of the cluster, but grain classes should only be referenced by the silos that will host them.
What I have tested, this is actually not true. The method parameter type is the problem, not the grain interfaces.

COCPORN

@COCPORN

Where is this documentation? I am curious.

Panu Oksala

@PanuOksala_twitter

https://dotnet.github.io/orleans/Documentation/clusters_and_clients/heterogeneous_silos.html

COCPORN

@COCPORN

Thanks! I'll look at it. I am sorry, I am unfamiliar with this feature, so I suspect I cannot really be helpful.

Where do you run the codegeneration?

Panu Oksala

@PanuOksala_twitter

we use the codegenerator.msbuild nuget package, which creates the files during build.

that package is referenced at least in the silo hosts

COCPORN

@COCPORN

Is it in the interface assembly, tho?

Thanks for the tip about this feature, BTW. It has passed me by completely.

Panu Oksala

@PanuOksala_twitter

Yes

burgerTo

@burgerTo

Hi,
i have a question regarding async running methods.

This is a examplecode of a GrainMethod:

public Task GrainMethod()
{
_anotherGrain = GrainFactory.GetGrain<IBasicMeasurementDevice>(guid);

_anotherGrain.AnotherGrainMethod().ContinueWith(task =>
{
    //modify private members in Grain;
}, TaskContinuationOptions.OnlyOnFaulted);

}

I want to run the "AnotherGrain" Method without awaiting for it. But if there occurs a exception, I want to modifie the state of the caller grain.
Could this solution lead to race conditions?

oleggolovkovuss

@oleggolovkovuss

Hello, does anyone know what should I use instead of
options.Invariant = "MySql.Data.MySqlClient"; //this was working with mysql 5.6
with MySql 8?

oleggolovkovuss

@oleggolovkovuss

actually the invariant seems fine, it doesn't like the creation of stored procedures now for some reason

oleggolovkovuss

@oleggolovkovuss

ok, my error was because of this line https://github.com/dotnet/orleans/blob/master/src/AdoNet/Orleans.Persistence.AdoNet/MySQL-Persistence.sql#L92 though the exception message was saying there's wrong syntax in a diffrent location omg, thanks all :D

Panu Oksala

@PanuOksala_twitter

@COCPORN I have an example project available that verifies this issue

Sergey Bykov

@sergeybykov

@COCPORN

@sergeybykov Yes, it was hyperbole from me. About the buckets, perhaps I've misunderstood how they work. It looks to just be sorted on expiry time, and then I suppose any grain that has activity moves itself to another bucket based on the time to live configuration. I could take a stab at implementing my suggestion, as I think the amount of work needed to have something that is useful (or can at least be evaluated for usefulness) isn't that big.

Yes, when a grain is invoked, it gets moved to the most recent bucket, unless it's already there. If you are willing to take a stab at it, we'll be happy to help.

@PanuOksala_twitter http://dotnet.github.io/orleans/Documentation/clusters_and_clients/heterogeneous_silos.html says: "All silos should reference interfaces of all grain types of the cluster, but grain classes should only be referenced by the silos that will host them." That also implies that any serializable types that could be sent as arguments should be referenced by all silos. Maybe worth adding such a statement to the doc to prevent confusion.

Sergey Bykov

@sergeybykov

@burgerTo I believe what you are trying to do will generally work, with all the potential implications of not awating an outgoing call - the caller grain might get deactivated before the response arrives, etc.

David Christensen

@OracPrime

Is there any documentation on custom placement strategies? There are the examples inside the Orleans repo, but I feel like I'm reverse engineering...

Reuben Bond

@ReubenBond

There aren't yet, @OracPrime - but they would be welcome :)

David Christensen

@OracPrime

Yes, wouldn't they :)

Barry Briggs

@barrybriggs

Ref the docs on external tasks: https://dotnet.github.io/orleans/Documentation/grains/external_tasks_and_grains.html I'm assuming from reading that I could use this method to call out from a grain to a very-long-running-task (like a Python interpreter training an ML model) and have it periodically call back into a grain?

(Just want to check before writing a bunch of code :-))

Reuben Bond

@ReubenBond

You can do that, @barrybriggs, yes. External tasks/threads can call back into grains. Certainly, it is advised to start such tasks using Task.Run.

Barry Briggs

@barrybriggs

thx!

Tom Nelson

@Zeroshi

@sergeybykov

I think it'd be better for us to chat thematically over Teams/Skype than for me to try to type the same thoughts, and then for us to go back and forth. A voice dialog is an efficient means of arriving to something. 🙂

That sounds great!

thank you!

Sergey Bykov

@sergeybykov

@Zeroshi What time zone are you in?

David Christensen

@OracPrime

Hmmm. Custom placement strategies appear to not as hard as at first thought (famous last words)

Sergey Bykov

@sergeybykov

@OracPrime They are not. Usually, the harder part is getting data necessary to make placement decisions (statistics, etc.).

Tom Nelson

@Zeroshi

@sergeybykov im on eastern

Veikko Eeva

@veikkoeeva

Learned a ton about designing and ordering business cards over the weekend. Things look so nice in RGB and quite more bland in CMYK. :P

Tom Nelson

@Zeroshi

I used to be a graphic designer. I can help with that

Panu Oksala

@PanuOksala_twitter

@sergeybykov ok, thanks

Veikko Eeva

@veikkoeeva

@Zeroshi I wonder... :) I'm currently actually looking a printing shop that can help me produce a pdf with the right color pallette information and help select choosing the right material (likely paper) and finishes. I've been thinking fluorescent green edge color and maybe similar components or metallic (just metallic or holograhic) elements on the surface. The current design is like

Where communities thrive

People

Repo info

Activity