Re: [akka-user] Split Brain?

ree in some sort of decentralized sorting criteria, for > that I use current time in millis + node address to make sorting of nodes > consistent among all other nodes. > > That way they don't need to query anyone to know who is the current active > node for an specific task.

[akka-user] Re: How to upgrade a mutable stateful Akka application in product without downtime ?

I have a somewhat similar issue I am dealing with. Our volume requirements aren't enormous, so I have separated processing of commands into two states 1. Save the command in an external database in a "Pending" status 2. Process the command. 3. Update the command to be completed when it is done.

Re: [akka-user] Split Brain?

Thanks, yea, guess it makes sense to have each member auto-down themselves... "If my cluster doesn't reach quorum size, then stop me" > > -- >> Read the docs: http://akka.io/docs/ >> Check the FAQ: >> http://doc.akka.io/docs/akka/current/additional/faq.html >>

Re: [akka-user] Split Brain?

Thanks Patrik... So, if I just put all nodes in the seed node list, then there would be no split brain on start, did I follow that correctly? I planned on implementing a simple resolver based on a static quorum size. Seems pretty simple to just have a cluster singleton that listens to cluster

[akka-user] Split Brain?

I noticed that 2.4.4 had some work on the Split Brain Resolver. Is this going to be OSS or remain commercial? I have recently seen the dreaded split brain, and need to implement a workaround. A shared quorum value should be sufficient for my purposes. Also, just curious, is there anyway that

Re: [akka-user] Distributed Data and Akka Cluster Sharding

to remove akka persistence would hint that ddata is not yet > covering that use case. > > On Thursday, 7 April 2016 11:18:56 UTC+1, Paul Cleary wrote: >> >> I do have this dependency in my build... >> >> "com.typesafe.akka" %% "akka-distributed-data-exp

Re: [akka-user] Distributed Data and Akka Cluster Sharding

I do have this dependency in my build... "com.typesafe.akka" %% "akka-distributed-data-experimental" % akkaV, On Thursday, April 7, 2016 at 6:16:40 AM UTC-4, Kealan Murphy wrote: > > From here > > >

Re: [akka-user] Distributed Data and Akka Cluster Sharding

Here is the config I am using... I am using the dynamodb akka persistence plugin right now and it works fine when it is in there. If I take out the persistence configuration all together then cluster sharding does not start up, I see this error: 16-04-07 01:06:14,514 [test] ERROR | [project-a

[akka-user] Re: Dedicated seed nodes for akka cluster

This looks interesting, has any work been done to use consul kv by any chance? Does this solution avoid the split brain? On Thursday, March 17, 2016 at 8:32:50 PM UTC-4, Rafał Krzewski wrote: > > There are ways to boot up Akka cluster without dedicated seed node: > whichever node comes up first

Re: [akka-user] Re: Dedicated seed nodes for akka cluster

Is there a recommendation (beyond using the split brain resolver which looks like you need a subscription?) for manual downing or managing split brain? On Saturday, April 2, 2016 at 9:45:10 AM UTC-4, Patrik Nordwall wrote: > > If you use auto-downing and that triggers because of a network partit

[akka-user] Distributed Data and Akka Cluster Sharding

I am using Cluster Sharding with the Distributed Data Module. I have remember entities on so that I can auto-start an entity during failover. Even with the distributed data module, it appears as though I _need_ to have an akka persistence journal or else my Actor system doesn't start. Is there

Re: [akka-user] Quite stdout when using the TestEventListener?

(EventFilter.info())) > > or a certain exception: > >system.eventStream.publish(Mute(EventFilter[MyException]("message"))) > > -Endre > > On Wed, Feb 17, 2016 at 1:50 PM, Paul Cleary > wrote: > >> I am using the akka testkit TestEventListener to

[akka-user] Re: Scaling actors

I use HA PROXY and have blue / green (active + dark) backends. This allows us to do a seamless traffic shift during an upgrade. It also allows you to test the upgrade prior to making it "live". For the app, it depends on what you want to do. If you are using DDD, CQRS and model domains, then

[akka-user] Quite stdout when using the TestEventListener?

I am using the akka testkit TestEventListener to check for exceptions. I have a number of test cases spread across the application that use it. My application.conf for src/test looks like: akka { loglevel = "INFO" loggers = ["akka.testkit.TestEventListener"] But now, I get a TON of output in

Re: [akka-user] Auto Restart actor when cluster node fails

la/cluster-sharding.html#Remembering_Entities > Have you seen that feature? :) > > -- > Cheers, > Konrad 'ktoso’ Malawski > Akka <http://akka.io> @ Typesafe <http://typesafe.com> > > On 9 February 2016 at 16:01:06, Paul Cleary (pcle...@gmail.com > ) wrote:

Re: [akka-user] Auto Restart actor when cluster node fails

> Konrad 'ktoso’ Malawski > Akka <http://akka.io> @ Typesafe <http://typesafe.com> > > On 9 February 2016 at 16:01:06, Paul Cleary (pcle...@gmail.com > ) wrote: > > I am using Cluster Sharding to manage certain actors. > > Is it possible that, when a node d

[akka-user] Re: Akka Persistent Actor Lifecycle

You can use Passivate, which will persist I believe after some period of unuse On Tuesday, February 9, 2016 at 9:57:05 AM UTC-5, Mahmoud Atef wrote: > > Hi, > > I'm new to AKKA world and I'm working on implementing EventSourcing using > the PersistentActors, but I got confused about the life-cy

[akka-user] Auto Restart actor when cluster node fails

I am using Cluster Sharding to manage certain actors. Is it possible that, when a node dies and the shards are re-distributed to other nodes in the cluster, to have the actor automatically start up? -- >> Read the docs: http://akka.io/docs/ >> Check the FAQ: >>>

[akka-user] Re: Loading Actor State from DB and Failures

Thanks a lot for all your help on this. I think I got this nailed. The BackoffSupervisor is sufficient, if not a little goofy to deal with. I am thinking that it would be better to create a supervisor strategy to emulate the backoff. I am unclear as to why Akka didn't make this out of the box

[akka-user] Re: Loading Actor State from DB and Failures

I found the issue. 1. Just throwing an exception seems to bypass the BackoffSupervisor (unfortunately), which immediately starts the restart loop for the actor (not using the backoff) which really stinks 2. The BackoffSupervisor does not do a normal akka *restart* on the actor. Instead, it *co

[akka-user] Re: Loading Actor State from DB and Failures

Something is not quite right. It seems like the "BackoffSupervisor" is not backing off. It is restarting immediately, as opposed to any kind of wait. Still investigating. Thanks for your help, I will post my solution once I get it behaving properly. Right now, I am trying to figure out why t

[akka-user] Re: Loading Actor State from DB and Failures

I just got this... When using the *BackoffSupervisor* for a child, calling *context.stop(self)* or *self ! PoisonPill* from *the child* will invoke supervision and force a restart. Therefore, when using the *BackoffSupervisor*, in order to terminate the child normally, you MUST do a *context.p

[akka-user] Re: Loading Actor State from DB and Failures

Might be simpler to review here...? https://gist.github.com/pcleary00/b5d02c3d4aef151c5a0f -- >> Read the docs: http://akka.io/docs/ >> Check the FAQ: >> http://doc.akka.io/docs/akka/current/additional/faq.html >> Search the archives: https://group

[akka-user] Re: Loading Actor State from DB and Failures

Here is a complete example that I have in my head. Job Manager -- 1 to many --> Job -- 1 to many --> Task // - DOMAIN MODEL sealed trait TaskStatus case object Pending extends TaskStatus case object Complete extends TaskStatus case class Task(id: String, status: TaskStatus =

[akka-user] Re: Loading Actor State from DB and Failures

Thanks, if you don't mind, I will code this out and post it in here? Will be more complete to show the 3 things working -- >> Read the docs: http://akka.io/docs/ >> Check the FAQ: >> http://doc.akka.io/docs/akka/current/additional/faq.html >> Searc

[akka-user] Re: Loading Actor State from DB and Failures

re: "The mechanism is a bit tricky", thanks goodness it isn't just me ;) Say I have a *JobSupervisorActor* that really manages work. He might get a message *StartJob(job)* for example. Handling might look like: 1. Save the job in the db 2. When job is saved successfully, spawn a *JobActor* chil

[akka-user] Re: Loading Actor State from DB and Failures

Wait, when a job finishes, it sends a poison pill to its *parent*? Or to its self? -- >> Read the docs: http://akka.io/docs/ >> Check the FAQ: >> http://doc.akka.io/docs/akka/current/additional/faq.html >> Search the archives: https://groups.googl

[akka-user] Re: Loading Actor State from DB and Failures

Thanks a lot for the discussion Guido, I am going to try and parse what you said here following the example. If any individual Task fails, the Job Actor would be responsible for handling it (by default restarting it / retrying it). If the Job fails, then some parent (lets call him the Job Superv

[akka-user] Re: Loading Actor State from DB and Failures

Thanks, yea, I like the Backoff Supervisor, very handy. I tinkered with that already for handling this kind of scenario. The next challenge that follows here is what to do when you are maintaining state in a database. Imagine you have a long running process / job that you are monitoring. You

[akka-user] Re: Loading Actor State from DB and Failures

nce as the mailbox has >> different requirements and the code will be less readable. >> >> HTH, >> >> Guido. >> >> On Friday, February 5, 2016 at 2:07:24 PM UTC, Paul Cleary wrote: >>> >>> I am trying to figure out the best way to load

[akka-user] Re: akka-persistence and serialization

Eric, The implications of java serialization are the following: 1. It is slow to serialize and deserialize relative to other means 2. You cannot run a mixed bag. If you have a type User that is version1, and then you change the User to be version2, if you serialize using something like akka per

[akka-user] Loading Actor State from DB and Failures

I am trying to figure out the best way to load actor state from a database. The actor could be a child actor of another. >From what I have found, something like this would work: class MyActor extends Actor with Stash { var data: Data = _ // start the actor in a loading state def recei

[akka-user] Loading Actor state from DB and failure

I want to be able to load state for an actor from a database. It appears as though a standard pattern is something like the following: class MyActor extends Actor{ } ss -- >> Read the docs: http://akka.io/docs/ >> Check the FAQ: >> http://doc.akka.io/docs/

Re: [akka-user] Handling Failures from Futures in an Actor

e actor so that it can affect the actor—and a message is the only > way to do that. “Unsavory” is by definition relative to your taste ;-) but > I can say that what you describe is certainly idiomatic actor code (apart > from the println—e ;-) ). > > Regards, > > Roland

[akka-user] Handling Failures from Futures in an Actor

I am trying to figure out the best way to handle a failure of a future in an Actor. My use case is interacting with a database, which as you can expect is asynch / Future based. Ideally, I want the actor to restart. Here is a relevant code snippet: def running: Receive = { case "do something

[akka-user] Handling Failures from Futures within Actors

I am trying to figure out the best way to handle failures from futures within an Actor. If I have an external resource that can throw an exception / return a Failure in a future, I really want the Actor to restart. Here is a relevant code snippet: def running: Receive = { case "do something"

Re: [akka-user] Testing actor failure from a future?

method synchronous in the trait to ease testing, then in your > actor you can wrap the synchronous call in a future. i like doing it this > way, it separates the computation part from the coordination part, and > makes both pieces simpler to test. > > -Michael > > On 02/01/16 17:22,

[akka-user] Re: Actor ask pattern always times out when calling an actor delegating work to its children

I agree with others, need to post some more code here. Also, how are you testing? I believe the TestKit and TestActorRef run single threaded. On Sunday, January 31, 2016 at 1:41:35 PM UTC-5, bangs wrote: > > I have an actor RequestConsumer (exposed over Camel netty http interface) > which cons

[akka-user] Testing actor failure from a future?

I have a child actor that I am trying to test using akka test kit. Pretty simple really: def receive = { case x => process(x).map(_.right) pipeTo sender } The process method yields a future. Now, I am trying to write a unit test (using the akka test kit), and nothing I seem to do captures t

Re: [akka-user] Akka Persistence with Multi-Clusters

’ll have to make sure that you don’t emit events to the > same log from different places—there can only be one running source of > truth for each persistenceId at any given time. > > Regards, > > Roland > > 22 jan 2016 kl. 14:52 skrev Paul Cleary >: > >

[akka-user] Akka Persistence with Multi-Clusters

Will Akka Persistence work if you have two different clusters pointing to the same data store? Imagine I have 2 data centers that point at the same database. If I have updates happening in both data centers at the same time, will akka persistence stomp all over the journal / snapshots? I know

[akka-user] akka persistence plugin upgrade help

advance, Paul Cleary -- >>>>>>>>>> Read the docs: http://akka.io/docs/ >>>>>>>>>> Check the FAQ: >>>>>>>>>> http://doc.akka.io/docs/akka/current/additional/faq.html >>>>>>>>>

Re: [akka-user] dynamodb snapshot plugin

to > avoid long recovery times when replaying accumulated events. > > /Patrik > > On Tue, Sep 29, 2015 at 3:01 PM, Filippo De Luca > wrote: > >> The dynamodb object are limited to 400K, implementing a snapshot plugin >> on that, maybe pointless. My 2cents. >

Re: [akka-user] dynamodb snapshot plugin

ted to 400K, implementing a snapshot plugin on > that, maybe pointless. My 2cents. > > On 28 September 2015 at 22:48, Paul Cleary > wrote: > >> Thanks Konrad, yes, I am using that plugin for my journal. >> >> Akka Clustering stores cluster state (singletons, shard

Re: [akka-user] dynamodb snapshot plugin

f in trouble while > implementing a journal. > > On Mon, Sep 28, 2015 at 10:59 PM, Paul Cleary > wrote: > >> Before I go and create a dynamodb snapshot plugin of my own, I wanted to >> see if anyone else had done this and maybe it is open sourced someplace (or >>

[akka-user] dynamodb snapshot plugin

Before I go and create a dynamodb snapshot plugin of my own, I wanted to see if anyone else had done this and maybe it is open sourced someplace (or is planned on being open sourced). Seems simple enough to write, but this question is much faster ;) Thanks! -- >> Read the docs: h

Re: [akka-user] akka cluster sharding issues with docker?

eads on this mailing > list that explain in detail why that is. > > Regards, > > Roland > > 20 sep 2015 kl. 02:12 skrev Paul Cleary >: > > Thanks, will check it out...but...ugh...I am on akka 2.3 > > On Saturday, September 19, 2015 at 9:06:20 PM UTC-4,

[akka-user] Re: akka cluster sharding issues with docker?

Is it possible there is some kind of issue using AtLeastOnceDelivery here? I am using at least once delivery between the cluster singleton and the cluster shard Here are some more logs that are happening on the seed node that is sending the message to the shard on the "other" node: -- so, it

[akka-user] Re: akka cluster sharding issues with docker?

To summarize, here is my setup: App ---> Cluster Singleton ---> Cluster sharded actor In the failure scenario above, this looks like: App(on seed) ---> Cluster Singleton(on seed) --> Cluster sharded actor (on node1, separate node) So, my functional tests work fine when everything is in the same

[akka-user] Re: akka cluster sharding issues with docker?

Thanks, will check it out...but...ugh...I am on akka 2.3 On Saturday, September 19, 2015 at 9:06:20 PM UTC-4, tigerfoot wrote: > > Paul, > > Try looking at my repo here: https://github.com/gzoller/docker-exp > > Check out the cluster branch. It shows how to use the new dual-binding > features in

[akka-user] Re: akka cluster sharding issues with docker?

-4, matheus...@gmail.com wrote: > > You need use another config to correctly bind container ip because it uses > a NAT network. See: > > http://doc.akka.io/docs/akka/snapshot/additional/faq.html#Why_are_replies_not_received_from_a_remote_actor_ > > > On Friday, Sept

[akka-user] Re: akka cluster sharding issues with docker?

I upped the logging levels, a little more info...it seems as though this is happening when attempting to send messages through cluster sharding. In the following log snippet, I see this "endpointWriter - received local message RemoteMessage: [null]" In this interaction, 172.17.0.27 is the "node

[akka-user] Re: akka cluster sharding issues with docker?

I am almost certain that this is the cause: 13:08:28.723UTC [test] DEBUG akka.contrib.pattern.ShardRegion akka.tcp://Denis@172.17.0.22:2551/user/sharding/Zone - Request shard [-10] home 13:08:28.724UTC [test] DEBUG a.s.Serialization(akka://Denis) akka.serialization.Serialization(akka://Denis)

[akka-user] akka cluster sharding issues with docker?

I am struggling to get akka clustering working with docker in a sane way. I can manage to get it working if I use --net="host" and use the local loopback (127.0.0.1), but if I try and just use the ips for docker containers, it seems like Akka remoting falls down somewhere. Here is my setup: - r

[akka-user] Akka Persistence Event Sourcing with Read Projections / Views?

I am thinking that the simplest thing is to maintain a separate persistent view that gives me query flexibility. The aggregate root is run via event sourcing. I try to keep the view in sync, but will update the view whenever the aggregate root starts and finishes recovery. Not sure if there

[akka-user] Akka Persistence Event Sourcing with Read Projections / Views?

I have been building out a POC using event sourcing in order to manage the state of an aggregate root. This works great. Say I have an aggregate root "Client". I might have thousands of clients in my system (or more). When the user hits my API to say "Get me a list of all clients in the syst

[akka-user] Pending Mailbox on Actor Failover in Cluster

When using Akka Clustering, if a JVM crashes, what happens to the pending messages in the actors on the failed system? -- >> Read the docs: http://akka.io/docs/ >> Check the FAQ: >> http://doc.akka.io/docs/akka/current/additional/faq.html >> Search

[akka-user] Re: Akka Http Gzip Json

I have gone back to Spray 1.3.2 for the time being as this is a stopper for me. I spent a lot of time trying to sort through this Akka Http issue and threw in the towel. On Tuesday, March 24, 2015 at 8:23:01 AM UTC-4, Paul Cleary wrote: > > One other note with this issue, it seems when

[akka-user] Re: Running Cluster Sharding Example

Answering my own question here, but I seem to have figured it out. Must have been a copy / paste error from the template into my project. Not sure exactly what it was, but I am up and running. On Tuesday, March 24, 2015 at 4:43:03 PM UTC-4, Paul Cleary wrote: > > I am trying to run the c

[akka-user] Running Cluster Sharding Example

I am trying to run the cluster sharding example activator template project. It looks like everything starts up ok. But, then I am getting the following error: > Lookup of shared journal at > akka.tcp://ClusterSystem@127.0.0.1:2551/user/store timed out I am using the code: https://github.com/

[akka-user] Re: Akka Http Gzip Json

One other note with this issue, it seems when I get it, the java process never dies, and starts hogging all of my CPU. Perhaps there is some kind of out-of-control looping that is going on? On Monday, March 23, 2015 at 9:26:06 AM UTC-4, Paul Cleary wrote: > > I am trying to put toge

[akka-user] Re: Akka Http Gzip Json

I think this maybe a bug. I can get this working with Spray 1.3.2, but it does not work with Akka Http 1.0-M4. Unless there is a different method in Akka Http 1.0 that I am missing? On Monday, March 23, 2015 at 9:26:06 AM UTC-4, Paul Cleary wrote: > > I am trying to put together a rout

[akka-user] Akka Http Gzip Json

I am trying to put together a route that does a decode with Gzip, and then uses an entity(as. With gzip decoding on, my route is not running. If I remove Gzip, then everything works fine. Here is the error I am getting: > Request was neither completed nor rejected within 1 second > (RouteTes

Re: [akka-user] Implicit value issue for FromRequestUnmarshaller in akka-http with spray-json

Ugh, I hit this same issue using 1.0-M4; you still need an implicit `ActorFlowMaterializer` :( On Sunday, March 15, 2015 at 7:29:23 PM UTC-4, James Mulcahy wrote: > > I think you need an in-scope implicit FlowMaterializer (or perhaps an > ActorFlowMaterializer, if you're on M3. I think that may

[akka-user] Re: Can Cluster Sharding work across applications?

nly mode or pass > messages from frontend to an actor running on backend which will forward > them to shard region. > > On Friday, March 20, 2015 at 10:54:15 PM UTC+3, Paul Cleary wrote: >> >> I have two halves of an application, a front-end that is busy with >> ingesti

[akka-user] Can Cluster Sharding work across applications?

I have two halves of an application, a front-end that is busy with ingesting data, and a backend for processing the data. Is it possible to have the front-end "forward" messages to the back-end using Cluster Sharding? I am interested in Cluster Sharding because I need to be able to have a sing

Re: [akka-user] Concurrency in creating actors in cluster / remote

28 PM UTC-4, Martynas Mickevičius wrote: > > Hello Paul, > > have you considered cluster sharding > <http://doc.akka.io/docs/akka/2.3.9/contrib/cluster-sharding.html>? > > On Sun, Mar 15, 2015 at 5:34 AM, Paul Cleary > wrote: > >> I have to create a

[akka-user] Concurrency in creating actors in cluster / remote

I have to create a short lived actor in a cluster, the actor is identified by a UUID. It is possible that I have 10 messages coming in for the same UUID at the same time on 10 *different* nodes. How would I create the actor once for that UUID, and ensure that all 10 messages get to the same ac

Re: [akka-user] Acceptable Limits for Actor Creation and Destruction?

for everyone's help, this is great. On Tuesday, January 27, 2015 at 5:27:08 AM UTC-5, Björn Antonsson wrote: > > Paul, > > The child will tell the Parent to supervise it by sending the Parent a > system message. > > B/ > > On 26 January 2015 at 19:10:50, Paul Cle

Re: [akka-user] Acceptable Limits for Actor Creation and Destruction?

nt and > child needs CPU time) > > On Mon, Jan 26, 2015 at 5:49 PM, Paul Cleary > wrote: > >> Victor, >> Yes, they are spawned under the same parent (same actual actor ref). The >> parent is "long-lived", i.e. intended to hang around for the lifetime o

Re: [akka-user] Acceptable Limits for Actor Creation and Destruction?

n if they are spawned under the same parent or not. > > On Mon, Jan 26, 2015 at 5:07 PM, Paul Cleary > wrote: > >> Thank your for the reply, >> >> These machines are typically 4 core / 8 GB machines with only our app >> running. The only JVM settings are CM

Re: [akka-user] Acceptable Limits for Actor Creation and Destruction?

your actors? > > B/ > > On 21 January 2015 at 14:35:53, Paul Cleary (pcle...@gmail.com > ) wrote: > > I have an application where we are creating and destroying "a lot" of > actors. The amount of actors that are created is tied to the volume of > requests o

[akka-user] Acceptable Limits for Actor Creation and Destruction?

I have an application where we are creating and destroying "a lot" of actors. The amount of actors that are created is tied to the volume of requests on the application. I am using FSM for the Actor in question here. I do not fully understand what is involved with the setup / teardown of an i

Re: [akka-user] Akka Streams - Chunked Encoding

5, rklaehn wrote: > > It seems to me that you are gzip-compressing each file individually. > So the result of your request will be the concatenation of multiple > gzipped files, which does not make any sense. > > You should get some data though. Have you tried testing with wget

[akka-user] Akka Streams - Chunked Encoding

I am trying to created a chunked response from a series of files on the file system. I am using the Test Server code as the basis. Here is my sample code: val requestHandler: HttpRequest ⇒ HttpResponse = { case HttpRequest(GET, Uri.Path("/"), _, _, _) ⇒ val dir = "/test-data/"

Re: [akka-user] Exploring Akka Streams and Akka Clustering

or Spark? It would be very good > feedback to know what to improve with akka streams. > > B/ > > On 4 November 2014 at 13:48:12, Paul Cleary (pcle...@gmail.com > ) wrote: > > Working on a project where we need to do processing of high-volume data > arriving via Kafka. > &g

[akka-user] Exploring Akka Streams and Akka Clustering

Working on a project where we need to do processing of high-volume data arriving via Kafka. We did a POC in Spark Streaming, but we are not yet ready for Spark. I was looking at using Storm, but the abstractions make me sad, at least compared to how you can pipeline in Spark. I am investigatin

Re: [akka-user] Actor system unexpected slow down at load?

aug 2014 kl. 02:37 skrev Paul Cleary >: > > I have an actor system that is setup using all of the akka defaults. > > Nothing in the system has a lot of latency, we are talking about sub > millisecond (20 microsecond) latencies. > > All traffic flows through a single

[akka-user] Actor system unexpected slow down at load?

I have an actor system that is setup using all of the akka defaults. Nothing in the system has a lot of latency, we are talking about sub millisecond (20 microsecond) latencies. All traffic flows through a single actor to downstream actors. This "main" actor just forwards the message onto chil

Re: [akka-user] Actor Selection for Remote Actors

options and understood the differences > and trade-offs. To make it clear, some comments inline... > > > On Thu, Aug 14, 2014 at 3:17 PM, Paul Cleary > wrote: > >> I am looking into using Akka Remoting, I am using akka 2.2.3. >> >> I have to communicate with a

[akka-user] Actor Selection for Remote Actors

I am looking into using Akka Remoting, I am using akka 2.2.3. I have to communicate with a bunch of remote actors. I have a few questions about the best way to do this: 1. Do I simply send messages through an Actor Selection itself (there is a tell on actor selection)? My concern is that this