[akka-user] Re: Scheduling long term processes in cluster

Hi Filippo, One idea (of many I guess): 1. Have a dispatcher for long running futures. 2. For each task create an actor and within that actor create a future that each time such task status changes it sends a message to its actor. 3. Query such actor to get statuses. The reason

[akka-user] Re: Akka remote small/many messages throughput discussion?

-receiver pair which will scale better. Again, I only suspect that's the answer but I want to double check. Regards, Guido. On Friday, March 18, 2016 at 11:13:19 AM UTC, Guido Medina wrote: > > Given the following scenario, which design would fit best: > > Assume I have 2 rece

Re: [akka-user] Re: Clarification on unreachable nodes in cluster

; > Hi Ben > > There are some subtle exceptions to the convergence rule. Unreachable > members with status Down or Exiting are not counted. > > /Patrik > > On Thu, Mar 17, 2016 at 11:25 PM, Guido Medina <oxy...@gmail.com > > wrote: > > In fact in the other node

[akka-user] Re: Node quarantined

sue. I am observing this behavior whilst the > cluster seems stable with no nodes being added/removed. I suspect that this > issue was first observed when I upgraded a different library that brought > in a new version of the netty library. > > On Tuesday, March 22, 2016 at 6:23:14 PM U

[akka-user] Re: Node quarantined

Hi Benjamin, You have nodes with predefined ports, one thing I have which eliminates that problem for these nodes is that only my seed node(s) have the port set, the rest will just get a dynamic and available port, making it get a different port when you do a rolling restart. I suspect you are

[akka-user] Re: Node quarantined

to quarantine nodes when they are re-joining using the same address, not sure if here or if it is an actual Git ticket. HTH, Guido. On Tuesday, March 22, 2016 at 9:22:00 PM UTC, Benjamin Black wrote: > > I see the same issue with 2.3.14. > > On Tuesday, March 22, 2016 at 2:00:15 PM

Re: [akka-user] Re: Node quarantined

Hi Benjamin, For what I could understand from the issue, this is happening only to nodes that rejoined the cluster under the same address (host and port) so I believe that setting akka.remote.netty.tcp.port = 0 should solve the problem in the meantime, Cheers, Guido. -- >>

Re: [akka-user] Re: Node quarantined

*Correction:* Set that only for non-seed nodes. On Wednesday, March 23, 2016 at 9:38:44 AM UTC, Guido Medina wrote: > > Hi Benjamin, > > For what I could understand from the issue, this is happening only to > nodes that rejoined > the cluster under the same address (host and

Re: [akka-user] Re: Clarification on unreachable nodes in cluster

I have a supervisor actor per node which is already subscribed to cluster events, I avoided using ask when the actor system is going down anyway and it is driven by the JVM shutdown hook, so instead I came up with this hack, maybe a bit ugly but it works: For supervisor actor which once it

Re: [akka-user] Re: Clarification on unreachable nodes in cluster

; > On Fri, Mar 18, 2016 at 6:47 PM, Guido Medina <oxy...@gmail.com > > wrote: > >> Won't happen, the micro-services are being sent Linux SIGTERM, which is >> why I'm hooking on the JVM shutdown. >> >> On Friday, March 18, 2016 at 5:26:01 PM UTC, Konrad Malaws

Re: [akka-user] Re: Akka HTTP Websockets Java with Actor

put here, Guido. On Friday, March 18, 2016 at 4:06:08 PM UTC, Konrad Malawski wrote: > > New Java API is comign in very large steps, I think the new one you'll > like a lot and come back to plain Akka :-) > > > On Mon, Mar 14, 2016 at 5:13 PM, Guido Medina <oxy...@gmail.com

Re: [akka-user] How to get parent of router from routee?

It is also possible to pass the grandpa actor via creator, that way it doesn't matter what member of the family you pass, it is always part of your creator and accessible to the actor. On Sunday, March 27, 2016 at 6:06:22 PM UTC+1, Konrad Malawski wrote: > > If it were me, I'd instead make this

Re: [akka-user] Re: Scheduling long term processes in cluster

ask and each of this actors are the ones that actually create the Future. That's the best I can come up in few minutes, Guido. On Friday, March 18, 2016 at 9:25:14 PM UTC, Guido Medina wrote: > > Remember amount of active actors != amount of running processes, you can > have 100 actors mo

Re: [akka-user] Akka remote communication roadmap?

working on this important area, and community > involvement is very much welcome. > > Cheers, > Patrik > > On Thu, Mar 3, 2016 at 6:29 PM, Guido Medina <oxy...@gmail.com > > wrote: > >> In the meantime I can't stay idle so I started digging inside Netty 3

[akka-user] Re: Using the Actor uid

*UID* = Unique Identifier just like: *UUID* = Universal Unique Identifier. On Friday, March 4, 2016 at 8:31:50 AM UTC, Tarisai Kunaka wrote: > > Hello > > I'd like to ask what exactly is an actor UID? > > >> -- >> Read the docs: http://akka.io/docs/ >> Check the FAQ:

Re: [akka-user] Akka remote communication roadmap?

In the meantime I can't stay idle so I started digging inside Netty 3 and at the moment I'm using my own Netty 3.10.6.Final (Not release yet) which replaces the Netty internal CHM (Was the one from JDK 5) There is a PR which most likely will be on 3.10.6.Final so be watching, I didn't bother to

[akka-user] Re: akka-cluster bandtwith issue

If your messages have medium to big size you could also enable Kryo compression, besides that at this moment there is nothing else you can do on top of Akka remote, see my last post here but don't add to it please as it has grown

[akka-user] Re: Actor restart and cluster re-announce

About a rolling restart pattern, this thread <https://groups.google.com/forum/#!topic/akka-user/0n-pxxjcJoI> has a nice discussion. On Friday, April 1, 2016 at 5:43:11 PM UTC+1, Guido Medina wrote: > > Many issues have fixed between 2.3.4 and 2.3.15 (released today), > includin

[akka-user] Re: Actor restart and cluster re-announce

Many issues have fixed between 2.3.4 and 2.3.15 (released today), including cluster fixes. I would recommend you to update to 2.3.15, that if you can't move to 2.4.3 (also released today) HTH, Guido. On Friday, April 1, 2016 at 5:36:52 PM UTC+1, Harold Poskanzer wrote: > > Hey, folks. I'm

Re: [akka-user] Can I use in-memory Actor state as a search engine ?

Hi Chelios, The problem you are solving is divided in two and I think it has been resolved before though it is quite complex but if you divide and conquer it might turn out to be easy. IMHO here are the main aspects of your problem: - Your data is distributed, each node with data will

Re: [akka-user] Can I use in-memory Actor state as a search engine ?

in Erlang, they have actors too, and they still have to use BloomFilters from Google. They all basically copies of the same paper which basically tells you ways to reduce data very fast using well known hashing techniques. Guido. On Wednesday, March 30, 2016 at 11:49:24 AM UTC+1, Guido Medina wro

Re: [akka-user] postStop doesn't get invoked after actor initialization failure

tructor fails. Which could be > said to be similar to preStart? > > -- > Cheers, > √ > On Mar 31, 2016 2:39 PM, "Endre Varga" <endre...@lightbend.com> wrote: > > > > On Thu, Mar 31, 2016 at 2:36 PM, Guido Medina <oxy...@gmail.com> wrote: > &

[akka-user] Re: Can I use in-memory Actor state as a search engine ?

If you want to have an actor per domain object handled by akka-sharding for read/write and after each write update the indexes affected by that object for an in-memory map reduce engine which again one exist per akka-shard node. That is...very do-able, you can save yourself sometime and use a

Re: [akka-user] postStop doesn't get invoked after actor initialization failure

- init; preStart (fails)?; postStop (fails)? > > The above contract is basically: "If initialization succeeds, preStart and > postStop are attempted to be called eventually, independently of thrown > exceptions" > > I proposed the second as it makes postStop a new line of def

Re: [akka-user] postStop doesn't get invoked after actor initialization failure

I was only thinking in a consistent life cycle for both. With the new behavior I think I will never use preStart again in Akka Typed. That will be the only way we the users make sure we have control on triggering conditions to execute postStop or not. Regards, Guido. -- >> Read

Re: [akka-user] postStop doesn't get invoked after actor initialization failure

ehavior of Akka Typed (which is still experimental) to be what I consider > more consistent now, but leaving untyped actors unchanged. Any objections? > > Regards, > > Roland > > 1 apr 2016 kl. 13:58 skrev Guido Medina <oxy...@gmail.com >: > > @Roland, > > Ind

Re: [akka-user] postStop doesn't get invoked after actor initialization failure

And the order would be: *new instance* -> *preStart()* -> *postStop()* - if started -> *preDestroy()* - if instantiated On Friday, April 1, 2016 at 1:42:16 PM UTC+1, Guido Medina wrote: > > I'll throw you an idea, what if you add another method, say preDestroy() > to matc

Re: [akka-user] postStop doesn't get invoked after actor initialization failure

the code is experimental and the change is highly desired > - the existing semantics are considered a bug > > -Endre > > On Thu, Mar 31, 2016 at 2:39 PM, Guido Medina <oxy...@gmail.com > > wrote: > >> I was exalting scenario with "legacy code life miserabl

Re: [akka-user] postStop doesn't get invoked after actor initialization failure

its parts are valid and started... But go ahead, make legacy code life miserable...the decision is yours. Try catch is there for a reason but somehow magic exists. Guido. On Thursday, March 31, 2016 at 1:15:27 PM UTC+1, drewhk wrote: > > > > On Thu, Mar 31, 2016 at 2:10 PM, Gu

Re: [akka-user] postStop doesn't get invoked after actor initialization failure

maintain the exceptional cases. On Thursday, March 31, 2016 at 1:23:39 PM UTC+1, Guido Medina wrote: > > so, say: > > Object a = null, Object b = null; > > preStart(): > > a = new something(); a.start(); *<= say I failed here.* > b = new somethingElse(); b start(); > &

Re: [akka-user] postStop doesn't get invoked after actor initialization failure

I was exalting scenario with "legacy code life miserable" meant: *Code that relied already on the current contract, no offense meant ;-)* On Thursday, March 31, 2016 at 1:35:24 PM UTC+1, drewhk wrote: > > > > On Thu, Mar 31, 2016 at 2:23 PM, Guido Medina <oxy...@gmai

Re: [akka-user] Can I use in-memory Actor state as a search engine ?

me. > > Apache Crunch looks great, may be there is a Scala client for this. I will > read up on it more. > > > > > On Wednesday, March 30, 2016 at 9:53:51 PM UTC+11, Guido Medina wrote: >> >> Even if you want to do it yourself you still have to reduce data fro

Re: [akka-user] Re: Scheduler overflow

CPU wise in today's CPUs Long operations and comparison are faster than with Integer (Integers are up cast to Long) but memory wise Long has 2 more bytes, GC wise I think there is more work to do with Integers than Long even if you end up consuming more longer term memory. That's getting down

Re: [akka-user] Re: Scheduler overflow

, Patrik Nordwall wrote: > > That is interesting. The scheduler will reject tasks scheduled too far > into the future, but you are not doing that. Please create an issue: > https://github.com/akka/akka/issues > > Regards, > Patrik > > On Thu, Apr 28, 2016 at 9:49 AM, Gui

Re: [akka-user] Re: Scheduler overflow

I'm not arguing about it, all I saying is: now in millis + times expressed as Integer limits it to 50 days now in millis + times expressed as Long limits it to 1000s of days. Guido. On Thursday, April 28, 2016 at 11:11:39 AM UTC+1, Konrad Malawski wrote: > > Yes, the reported issue is might be

[akka-user] Re: Surprising Akka Remoting behaviour: messages are delivered to (local) system with unresolvable hostname

Did you look at the Docker documentation for Akka remote configuration? http://doc.akka.io/docs/akka/current/java/remoting.html#Akka_behind_NAT_or_in_a_Docker_container HTH, Guido. -- >> Read the docs: http://akka.io/docs/ >> Check the FAQ: >>

Re: [akka-user] Re: Akka remote thread pools tuning.

goals, and how do you know that they are > met? > > On Wed, Apr 27, 2016 at 2:44 PM, Guido Medina <oxy...@gmail.com > > wrote: > >> I haven't deployed to servers so I cannot really tell, I'm afraid testing >> in my case won't give me much information because of lack of CPUs

Re: [akka-user] Re: Akka remote thread pools tuning.

It would at been nice to have such things documented at least at reference.conf Many thanks anyway, I'll try to figure it out. Guido. On Wednesday, April 27, 2016 at 2:06:01 PM UTC+1, √ wrote: > > > > On Wed, Apr 27, 2016 at 2:59 PM, Guido Medina <oxy...@gmail.com > > wro

[akka-user] Akka remote thread pools tuning.

Hi, I'm trying to tune the Netty client/server and remote dispatcher threads, assume I have the following: - There is a total of 5 servers. - Each server has 8 CPUs. - Each server is hosting 2 micro-services. I'm trying to tune Akka remote in a way that it is kind of ideal for that

Re: [akka-user] Re: Akka remote thread pools tuning.

s your measurements show? > > On Wed, Apr 27, 2016 at 2:32 PM, Guido Medina <oxy...@gmail.com > > wrote: > >> Should remote dispatcher ideally have the sum of client + server threads >> or min (CPUs, client + server threads) ? >> >> --

Re: [akka-user] Akka 2.4.5 broken actor creators

Is it safe to say it will be fixed in the next release? i.e.: 2.4.6? On Tuesday, May 17, 2016 at 9:28:58 AM UTC+1, Konrad Malawski wrote: > > The change is this: https://github.com/akka/akka/pull/20488 > > On Tue, May 17, 2016 at 9:50 AM, Guido Medina <oxy...@gmail.com > >

[akka-user] Akka 2.4.5 broken actor creators

My creators are not working with Akka 2.4.5, they have been working up to now, My actor creators are either static classes or standard classes but the methods that use the creator are not static: java.lang.IllegalArgumentException: cannot use non-static local Creator to create actors; make it

Re: [akka-user] Akka 2.4.5 broken actor creators

16 at 8:44:03 AM UTC+1, Konrad Malawski wrote: > > When talking about "that code does not work", please paste the code under > discussion. > > On Tue, May 17, 2016 at 9:39 AM, Guido Medina <oxy...@gmail.com > > wrote: > >> My creators are not working w

Re: [akka-user] Akka 2.4.5 broken actor creators

such a Creator and > we obviously didn't have test coverage for your way. I will investigate it > and report back. > > Regards, > Patrik > > On Tue, May 17, 2016 at 9:50 AM, Guido Medina <oxy...@gmail.com > > wrote: > >> I didn't pasted the code because I wasn

Re: [akka-user] Akka 2.4.5 broken actor creators

Yeah, agreed, I deleted that comment once realized the complexity of the problem. On Tuesday, May 17, 2016 at 10:15:30 AM UTC+1, Patrik Nordwall wrote: > > > > On Tue, May 17, 2016 at 10:32 AM, Guido Medina <oxy...@gmail.com > > wrote: > >> Is it safe to say it wil

Re: [akka-user] Akka 2.4.5 broken actor creators

Found the problem, an actor creator constructor need to be preferably public, even if the code compiles in Java, under some complex circumstances the class loader cannot see the constructor, in the reproducer posted on the issue https://github.com/akka/akka/issues/20537 I made it private to

Re: [akka-user] Implementing ACID style processing using Akka?

I developed a whole application using Akka + Java 8, for transactions I had no issues using a database pool and mapping with EclipseLink, my DAOs take care of initiating a new transaction or joining an existing one, the DB transactions problem I don't see any other way of handling, it shouldn't

Re: [akka-user] Implementing ACID style processing using Akka?

What Roland said is also very important, some operations can be scaled in a way. Eventually for the specific problem you are solving you should to scale depending under what category such activity belong to: - Shard-able. - Load balance-able. - Active - Passive (Singleton, can only one exist in

Re: [akka-user] Split Brain?

It might be also possible to use some sort of promotion algorithm, say, last in of a "role-X" takes over, the old know will see it and demote itself. All you need to do that is have one actor per node subscribed to cluster events, that;s how I do my promotion of active-passive nodes and it

Re: [akka-user] Split Brain?

> brain). > > Honestly, no idea how that side should be done. > > Step one would be just to shut down the node. > > Step two would be more graceful, of removing itself from the current > cluster, and figuring out how to join to the good cluster. > > On Thur

Re: [akka-user] Re: Node quarantined

Hi Ben, As my experience goes Netty 3 doesn't get much love, issues are barely fixed, like I mentioned before I'm running my own Netty 3.10.6 built internally, also; 3.10.0 is not even a good version, if you want force your version to 3.10.5.Final until they release 3.10.6.Final which has nice

[akka-user] Re: How to handle slow actors and bulkheading

gt; > > On Tuesday, June 28, 2016 at 10:30:38 AM UTC+2, Guido Medina wrote: >> >> There are community contributions that can poll messages from a JMS >> queue, then you could send these messages to a worker actor, >> such actor then can forward that message to ano

Re: [akka-user] Re: How to handle slow actors and bulkheading

Not to worry, everybody is happy now ;-) On Tuesday, June 28, 2016 at 10:51:30 PM UTC+1, Gavin Baumanis wrote: > > > Guido > > It isn't lost on me that this is the Akka mailing list. I too am here to > learn about Akka. > > I take onboard everything you say. But it was only a suggestion of an

[akka-user] Re: Node rejoin seed node failure after seed node restart

*Copy & paste from the other thread* Your nodes will re-join if you keep at least 1 seed node up while restarting other nodes, try having 2 seed nodes. In production don't ever relay on only 1 seed node, have at least 2. HTH, Guido. On Friday, July 8, 2016 at 11:44:52 AM UTC+1, 谌浏洋 wrote:

[akka-user] Re: Node rejoin seed node failure after seed node restart

You posted this question in another thread, please go there and look. On Friday, July 8, 2016 at 11:44:52 AM UTC+1, 谌浏洋 wrote: > > Sorry for my poor English > > I started two node A(seed) B, Both A and B can communication to each other. > > When I restart node A. node B cannot join A anymore. (No

Re: [akka-user] Re: node cannot automatically rejoin Cluster after seed node restarted

Your nodes will re-join if you keep at least 1 seed node up while restarting other nodes, try having 2 seed nodes. In production don't ever relay on only 1 seed node, have at least 2. HTH, Guido. On Friday, July 8, 2016 at 12:37:49 PM UTC+1, Patrik Nordwall wrote: > > When you restart node A

[akka-user] Re: Akka v2.3.14 with Java 7 - TLS configuration

To be honest and fair, TLS configuration is a messy "Java" subject, it has nothing to do with Akka so I believe you should search for Java TLS configuration instead, I know there is a JVM parameter where you indicate the current JVM which keystore should use but I don't know any details on how

Re: [akka-user] how to do a rolling cluster upgrade with marathon?

Hi Bert, What Martynas mentioned is the key to have a successful rolling restart, for production you need more than one seed node, I would even start my rolling restart with the seed nodes, it shouldn't make a difference but it will help the new started nodes not to rejoin seed nodes. There is

Re: [akka-user] Re: Is this a viable paradigm for PerRequest actors on Synchronous system?

that it is needed. > > Cheers, > Patrik > > On Wed, Jun 29, 2016 at 11:50 PM, Guido Medina <oxy...@gmail.com > > wrote: > >> Even if the cost of creating new actors is low, there is still a cost, >> its lifecycle for example has to be ran per actor

Re: [akka-user] Re: Is this a viable paradigm for PerRequest actors on Synchronous system?

with legacy code. On Thursday, June 30, 2016 at 9:17:35 PM UTC+1, Guido Medina wrote: > > That depends on the supervisor strategy, funny thing that for one my > clients with lots of legacy code I had to use a CompletableFuture because > the framework I'm introducing there has no of

[akka-user] Re: Is this a viable paradigm for PerRequest actors on Synchronous system?

But it can be improved; instead of creating and destroying actors per Future why not just create a pool of actors (or workers if you will), the recommended amount of concurrent actors you want is usually a factor of CPUs, say CPUs x N where N in [1..4], of course, and if these actors use legacy

[akka-user] Re: Is this a viable paradigm for PerRequest actors on Synchronous system?

akka.routing.SmallestMailboxRoutingLogic *Source:* http://doc.akka.io/docs/akka/2.4.7/java/routing.html -- >> Read the docs: http://akka.io/docs/ >> Check the FAQ: >> http://doc.akka.io/docs/akka/current/additional/faq.html >> Search the archives:

[akka-user] Re: Is this a viable paradigm for PerRequest actors on Synchronous system?

two types of "faster" and "cheaper" mailboxes for worker's like actors, like: https://gist.github.com/guidomedina/ddd729492c8fb8148032 That way my workers are always ready to process whatever. Just my opinion though, Guido. On Wednesday, June 29, 2016 at 10:41:52 PM UTC+1

[akka-user] Re: Is this a viable paradigm for PerRequest actors on Synchronous system?

Here is how to create your Router programmatically: final List actors = ... Router router = new Router(new RoundRobinRoutingLogic(), actors.stream().map(ActorRefRoutee::new). collect(toCollection(() -> new ArrayList<>(value.size(; In this case I'm using RoundRobinRoutingLogic,

[akka-user] Re: Is it possible to increase the number of serialization threads?

Try the following, these values work great, you can try higher but I don't think it is going to help because the most of the bottleneck is because of design, a problem related with how akka-remote is designed around Netty, but that will change soon: https://github.com/akka/akka-meta/issues/22

[akka-user] Re: Is it possible to increase the number of serialization threads?

. On Saturday, July 2, 2016 at 12:27:28 PM UTC+1, Guido Medina wrote: > > Try the following, these values work great, > you can try higher but I don't think it is going to help because the most > of the bottleneck is because of design, > a problem related with how akka-remote is design

[akka-user] Re: Is it possible to increase the number of serialization threads?

pool-size-factor = 1 pool-size-max = 8 } } } On Saturday, July 2, 2016 at 12:31:12 PM UTC+1, Guido Medina wrote: > > Didn't read the serialization part, must be that I need my coffee, Kryo is > very well optimized, you don't know if is going to help until you test it. &g

Re: [akka-user] Is it possible to increase the number of serialization threads?

serializer with its Kryo plus other things. You can see the relevant Akka Kryo code for the pool in this commit/diff: https://github.com/romix/akka-kryo-serialization/commit/045cd27dfd01c2c41ab7c64bf6e25a63b3fd8e42 Guido. On Sunday, July 3, 2016 at 1:47:45 PM UTC+1, Guido Medina wrote: > >

Re: [akka-user] Is it possible to increase the number of serialization threads?

I have mimic-ed your configuration and corrected some errors, also added Kryo if you want to give it a chance with the configuration I believe will do best. I default the "java" serializer to Kryo, that way, everything that inherits "Serializable" will use Kryo, also, I list every class that I

Re: [akka-user] Is it possible to increase the number of serialization threads?

il you don't call resolve() HTH, Guido. On Sunday, July 3, 2016 at 12:57:21 PM UTC+1, Guido Medina wrote: > > I have mimic-ed your configuration and corrected some errors, also added > Kryo if you want to give it a chance with the configuration I believe will > do best. > I d

Re: [akka-user] Is it possible to increase the number of serialization threads?

t; Anyway, kryo will scale vertically, I suppose. Will kryo use more threads > than I'm using right now? > > Thanks again for your help and knowledge. > > On Sun, Jul 3, 2016 at 2:13 PM, Guido Medina <oxy...@gmail.com > > wrote: > >> I meant to say *"anythi

Re: [akka-user] Is it possible to increase the number of serialization threads?

ware.kryo.serializers.CollectionSerializer.read(CollectionSerializer.java:96) > at > com.esotericsoftware.kryo.serializers.CollectionSerializer.read(CollectionSerializer.java:22) > at com.esotericsoftware.kryo.Kryo.readObject(Kryo.java:682) > at > com.esotericsoftware.kryo.serializers.ObjectField.read(ObjectField.java:106) > ... 32 mo

Re: [akka-user] Is it possible to increase the number of serialization threads?

nt out >> automatically registered classed, which is very helpful. >> >> I'm working on it. >> >> Thanks again for you patience. >> >> /Eduardo >> >> On Sun, Jul 3, 2016 at 11:53 PM, Guido Medina <oxy...@gmail.com >> > wrote: >> >

[akka-user] Re: Can I use Akka for such kind of architecture?

Hi Christian, Another possibility is to use Vert.x for the HTTP (with or without websockets) embedded with Akka if you find Akka Streams overwhelming which is what I have done, I have websockets via Vert.x and each socket connection creates an actor which basically have a mailbox (a queue)

[akka-user] Re: Cluster on different nodes minimally sharing code

+ParallelRefProcEnabled On Sunday, February 26, 2017 at 12:42:22 PM UTC, Guido Medina wrote: > > Hi Klaus, > > I have another example for you, I have an application in production > running among 15 JVM

[akka-user] Re: Cluster on different nodes minimally sharing code

Hi Klaus, I have another example for you, I have an application in production running among 15 JVMs, it is in Java 8 + Akka cluster (2.4.latest), The project itself is a Maven project with modules, the following are important aspects of my design: - each module has a different role and is

[akka-user] Re: how to load external configuration file in akka 2.0 ?

Hi Sergey, I have micro-services that run as Java main and the passed command line parameter is where my config file is located, I also fallback to an internal config (static.conf) which is part of a jar file, example below: final Config config = ConfigFactory.parseFile(new File(args[0])).

[akka-user] Re: Pattern for ensuring fault tolerant PARENT actors

*Solution 1:* Akka persistence (the parent only) with sharding can solve that problem for you, basically the parent actor will be re-instantiated somewhere else and resume from its last state and replay any message that follows. *Solution 2:* I used a similar concept where a type of parent is

Re: [akka-user] akka serialization - message boundary

Hi, You can also try Akka Kryo serialization extension, it has proven to be very efficient so far so you don't have to write your own serialization. Kryo offers you automatic, manual (by you listing explicitly) registration of the classes and both. It is a very matured extension and its backed

[akka-user] Re: Using distributed-pub-sub across DataCenters

Short answer is no, no, no, you need to interconnect the DCs using some recoverable mechanism otherwise your cluster will suffer. In other words, don't form Akka clusters among multiple DCs. HTH, Guido. On Wednesday, September 7, 2016 at 12:27:54 PM UTC+1, Guy Sela wrote: > > Hi, > > I saw

Re: [akka-user] Akka Multithreading on Akka-Cluster

Messages can be slow when being sent remotely from node A to node B, current Akka remote will give you a top of 100k msg/sec best scenario and that depending on the message size, there is a new Akka remote (Akka artery which is a rewrite of Akka remote) on the way so that shouldn't discourage

[akka-user] Re: Are messages ordered in activator hello akka java8 tutorial?

In the introduction chapter of Akka this is explained in details, how Akka takes care of it, I suggest you go there to learn the exact details: http://doc.akka.io/docs/akka/2.4.10/general/message-delivery-reliability.html#Discussion__Message_Ordering HTH, Guido. On Tuesday, September 6, 2016

Re: [akka-user] Akka Multithreading on Akka-Cluster

What do you suggest to focus our efforts on? > Which of this two arguments can effectively improve the system efficiency? > > - Passing from Remote to Artery? > - Remove Java serializer and use another one? > > Thanks, > Looking forward to hear you. > Silvio > >

Re: [akka-user] Re: Are messages ordered in activator hello akka java8 tutorial?

By induction you can conclude that messages order is guaranteed when sending sequentially (in the same thread) by using the following assertions: - Messages sent to any local actor go immediately to their destination inbox hence order is preserved because of sequential execution - JMM

[akka-user] Re: research about concurrency bugs

of, maybe too simple for your study but anyways, hope somehow it can lead you to what you need/want. Guido. On Wednesday, September 7, 2016 at 12:28:19 PM UTC+1, Guido Medina wrote: > > Hi Carmen, > > The following might not have a specific example but will probably help you > se

Re: [akka-user] Akka Multithreading on Akka-Cluster

g in Java and not in Scala) > Thanks Guys, > Silvio > > Il giorno mercoledì 7 settembre 2016 17:52:39 UTC+2, √ ha scritto: >> >> IMO Artery won't help much if using Java Serialization. >> Java Serialization is notoriously slow. >> >> On Wed, Sep

[akka-user] Re: Akka Multithreading on Akka-Cluster

And newest Akka Kryo doesn't require pool-size as that's a java.util.ConcurrentLinkedQueue by default. I contributed that new mechanism to use ConcurrentLinkedQueue for the pool or your own implementation (See KryoQueueBuilder example in their doc). On Monday, September 5, 2016 at 10:22:28 AM

Re: [akka-user] Akka Multithreading on Akka-Cluster

gt; "com.test.common.LoginRequest" = 20 >> >> >> } >> >> >> classes = [ >> "com.test.common.LoginRequest" >> >> ] >> } >> >> #akka.actor.serialize-messages =

Re: [akka-user] Akka Multithreading on Akka-Cluster

Also, what ID strategy are you using? I see none configured so it it is using the default, try setting it like I posted in the example which is "automatic" On Friday, September 9, 2016 at 5:26:00 PM UTC+1, Guido Medina wrote: > > Are you sharing the same conf file? if you re

Re: [akka-user] Akka Multithreading on Akka-Cluster

Hi Silvio, The following are the only things I can recommend you: - You have a short time to deliver? Why bother now with Artery if that's the least of your problems? Your performance issue has nothing to do with Akka Cluster or Akka Remote. - Why use a future? Your future is

Re: [akka-user] Enjoying Akka HTTP performance

If you want to test Netty indirectly and have some reasonable API in the middle try comparing with Vert.x 3. That way you are not just comparing against Netty but against a framework that is built on top of Netty, has a promising HTTP API and is matured (also faster than NodeJS of course):

Re: [akka-user] Akka Multithreading on Akka-Cluster

classes? If the latter then I believe it should also apply to automatic listed classes to register. Guido. On Friday, September 9, 2016 at 8:36:54 PM UTC+1, Justin du coeur wrote: > > On Fri, Sep 9, 2016 at 2:51 PM, Guido Medina <oxy...@gmail.com > > wrote: > >> T

Re: [akka-user] Akka Multithreading on Akka-Cluster

y, September 9, 2016 at 5:56:16 PM UTC+1, Justin du coeur wrote: > > On Fri, Sep 9, 2016 at 12:27 PM, Guido Medina <oxy...@gmail.com > > wrote: > >> Also, what ID strategy are you using? I see none configured so it it is >> using the default, try setting it li

Re: [akka-user] Akka Multithreading on Akka-Cluster

that is what i need. > > Thanks Silvio > > Il giorno mercoledì 7 settembre 2016 17:50:44 UTC+2, Guido Medina ha > scritto: >> >> Artery is at in milestone 3 (m3), you are welcome to use as I believe it >> is in an extremely good shape but I would pursue bo

[akka-user] Re: Using distributed-pub-sub across DataCenters

elaborate on this? > If I were to just use AKKA for point-to-point communication between multi > sites, would there be any problem there? > > On Wednesday, September 7, 2016 at 2:34:09 PM UTC+3, Guido Medina wrote: >> >> Short answer is no, no, no, you need to interconnect t

[akka-user] Re: thread-pool-executor with fixed-pool-size clarification

Also, I believe that it should be consistent, I think Java fixed thread pool factories will give you a thread pool that expire the threads after 1 minute being idle, should be the same *-consistency-* for Akka? On Monday, October 3, 2016 at 4:20:01 PM UTC+1, Guido Medina wrote: > >

Re: [akka-user] Re: Problem with clean shutdown using artery in cluster

Hmmm, maybe suggest Aeron guys to add an option to disable shutdown hooks, I know the same is possible for Log4j2 which is why my shutdown code fragment does once Akka system has been terminated. On Mon, Nov 28, 2016 at 11:05 AM, Patrik Nordwall wrote: > Thanks.

[akka-user] Re: Akka , priority order events

PriorityMailbox to the aid: http://doc.akka.io/docs/akka/2.4.14/java/mailboxes.html#PriorityMailbox HTH, Guido. On Sunday, November 27, 2016 at 10:00:44 PM UTC, adrian...@gmail.com wrote: > > I have heard a lot of good things about Akka. I have a stream of price , > and I need to create two

[akka-user] Re: Hashing Router vs Actor System with childs

If you have one actor per device why do you need a pool? I believe you are trying to use many *-and well known good-* concepts together which might over-complicate things, You could simply use one actor per device which will hold its own state and make it expire after some time being idle, such

[akka-user] Re: Problem with clean shutdown using artery in cluster

I had the same problem of members not leaving correctly, all because Aeron uses a JVM shutdown hook so your application cannot use also a shutdown hook, I had to switch to a provided SIGTERM class which catches the Ctrl + C or kill signal: public static void registerShutdownHook(ActorSystem

<    1   2   3   4   >