Re: [akka-user] SSL Support of akka-http(0.9)

Thank you for your reply. I'll follow those tickets. 2014年11月8日土曜日 0時56分16秒 UTC+9 Martynas Mickevičius: > > Hello Zirou, > > I am afraid this is still not possible also with the just released 0.10. > > You can follow progress on SSL in these > tickets

[akka-user] Anticipating Akka Persistence

It must be frustrating for the Akka team to field questions when the solutions will be so much more elegant with Akka Persistence and Reactive Streams :) I have an aggregate which needs to respond to events by updating its own persistent state (using Slick) and firing a domain event. I believe th

[akka-user] [akka-stream-experimental 0.10] FlowGraphBuilder.connect

Hi, I've been experimenting with merging multiple PartialFlowGraphs into a single FlowGraph, but have run into some difficulty. In the short example attached, I attempt to join two compatible graphs, where graphA.OUT is an UndefinedSink[Int], and graphB.IN is an UndefinedSource[Int]. I though

Re: [akka-user] synchronous lookups within an actor receive

OK thanks Martynas, I'll take a look further On Monday, 10 November 2014 14:11:10 UTC+2, Martynas Mickevičius wrote: > > Hi Brian, > > If everything needed to continue processing the request will be in the > answer to the ID lookup, then you can pipe >

Re: [akka-user] Re: Cluster node reconnects

On Mon, Nov 10, 2014 at 5:43 PM, Behrad Zari wrote: > The funny and bad thing is that when I tested my code today it was > working!!! (as I said in my previous post it also was working at start but > lately I couldn't get it working) > I'm confused since I haven't changed anything related to this

Re: [akka-user] Re: Cluster node reconnects

The funny and bad thing is that when I tested my code today it was working!!! (as I said in my previous post it also was working at start but lately I couldn't get it working) I'm confused since I haven't changed anything related to this :( So, Am i missing a bit of change mine, or it could it d

Re: [akka-user] Sharding problem when restarting Cluster

sure, no problem. On Monday, November 10, 2014 4:35:24 PM UTC, Patrik Nordwall wrote: > > > > On Mon, Nov 10, 2014 at 5:18 PM, Richard Bowker < > mechajoh...@googlemail.com > wrote: > > Hi Patrik, unfortunately not. In fact its only happened once to me so far > so may be a difficult one to repro

Re: [akka-user] Sharding problem when restarting Cluster

On Mon, Nov 10, 2014 at 5:18 PM, Richard Bowker < mechajohntravo...@googlemail.com> wrote: > Hi Patrik, unfortunately not. In fact its only happened once to me so far > so may be a difficult one to reproduce. > > I will of course get back to you if I can find a trigger. > Thanks. That is problem

Re: [akka-user] Sharding problem when restarting Cluster

Hi Patrik, unfortunately not. In fact its only happened once to me so far so may be a difficult one to reproduce. I will of course get back to you if I can find a trigger. Tackar! On Monday, November 10, 2014 4:04:18 PM UTC, Patrik Nordwall wrote: > > Hi Richard, > > That is not good. We have

Re: [akka-user] Sharding problem when restarting Cluster

Hi Richard, That is not good. We have seen similar issue a few times and tracked it down to bugs in the journal implementations. It will happen when events are replayed in the wrong order. Is there a way we can reproduce this? Regards, Patrik On Mon, Nov 10, 2014 at 2:42 PM, Richard Bowker < me

Re: [akka-user] Re: How expensive is the creation of an actor system? Is it suitable to create one on each request?

Hi Johannes, You're looking for a ServletContextListener, a quick googling turned this up, it looks OK, but YYMV: http://blog.nemccarthy.me/?p=295 On Mon, Nov 10, 2014 at 4:05 PM, oem oem wrote: > Hi Anders, > > thanks for your reply. > > As I understand: To get request processing working on J

[akka-user] Re: How expensive is the creation of an actor system? Is it suitable to create one on each request?

Hi Anders, thanks for your reply. As I understand: To get request processing working on Java-Appservers I will need a servlet to process the requests. So the servlet will be called for every request. As I see that as long as I don't want to keep the actor-system in the session I will need to

[akka-user] Re: Akka, Camel and RabbitMQ - Asynchronously acknowledging specific message

@Tal, You may be able to use https://github.com/sstone/amqp-client in place of akka-amqp. -Todd On Sunday, November 9, 2014 3:43:30 AM UTC-5, Tal Pressman wrote: > > Hi, > > I am trying to create a persistent actor that consumes messages from a > RabbitMQ queue, persists them asynchronously, a

Re: [akka-user] Re: Cluster node reconnects

Hi again, My hypothesis of why the node was marked as REACHABLE was wrong. The cluster heartbeat replies include the UID and replies from wrong incarnation are ignored. I have created a test that simulates the scenario as I have understood it. It behaves

Re: [akka-user] Getting the Result from Master Actor

Thanks Björn, I finally got it sorted through the use of Futures. Cheers On Tuesday, 4 November 2014 13:15:18 UTC+1, Björn Antonsson wrote: > > Hi Syd, > > On 4 November 2014 at 11:45:40, Syd Gillani (s.zeesha...@gmail.com > ) wrote: > > Hi, > > I am quite new to Akka (so it could be quite a nai

[akka-user] Re: How expensive is the creation of an actor system? Is it suitable to create one on each request?

Hi Why do you have to create the actor system on each request, and not simply create a new actor? I did not understand the following: "create an actor system for the request to get the parallelism working." Kind regards, Anders kl. 14:30:21 UTC+1 mandag 10. november 2014 skrev oem oem følgen

Re: [akka-user] Sharding problem when restarting Cluster

I have had seen a similar problem when restarting nodes in a cluster using sharding. after restarting, the node with the shard coordinator went into an infinite error loop. I was using akka 2.3.6 and "com.github.krasserm" %% "akka-persistence-cassandra" % "0.3.4" as the journal/persistence sto

[akka-user] How expensive is the creation of an actor system? Is it suitable to create one on each request?

Hi, I'm pretty new to Akka and would like to build a kind of resource-loader with it. On a HttpRequest I would like to gather async informations from different sources and return them aggregated. The Problem is that since this application has to run in a classical Java-Appserver like JBoss I'll

Re: [akka-user] synchronous lookups within an actor receive

Hi Brian, If everything needed to continue processing the request will be in the answer to the ID lookup, then you can pipe the future which you'll get while calling Reactive Couchbase APIs to the same actor and continue reque

[akka-user] Akka clusterclient Association Error / ClassNotFoundException when serialising response

Hi Everyone! I have been trying to implement akka clustering and sharding with persistent actor, i seperated the projects in 3 projects (*api, backend, and frontend*). I started 2 backend nodes with different ports, and one frontend cluster. I use cluster client in the frontend side, the comman

Re: [akka-user] Right Place To Instantiate Persistence View

Hi, a good idea is to use postStop to clean up everything that was created in preStart. For example scheduled tasks or PersistentView actors, etc. On Wed, Nov 5, 2014 at 9:19 AM, Prakhyat Mallikarjun wrote: > Hi Björn, > > Thanks for your inputs. > > Sorry its not "postStartup" method, we are c

[akka-user] Simplest possible setup wich allows system restart of individual nodes

Dear group In a project where performance is not an issue (yet, and possibly never), I am thinking about the following cold-standy solution to enable us to restart nodes without loosing state: One node is running with Persistent Actors saving to a database. When that node goes down, the other

[akka-user] synchronous lookups within an actor receive

Hi all I need help in designing an actor whose request needs to call a lookup on an ID before it can continue its request 'flow'. I was thinking the lookup could be done via another actor (against couchbase so it can be reactive), but I'm unsure how that call can be neatly returned into the ori