I'll test this out when I have "trunk" set up rather than the the 0.6
release.
> Here is a complete running sample (http://github.com/jboner/akka/blob/master/akka-core/src/test/scala/Se...
--
Jonas Bonér
twitter: @jboner
blog: http://jonasboner.com
work: http://scalablesolutions.se
code: http://github.com/jboner
code: http://akkasource.org
also: http://letitcrash.com
--
You received this message because you are subscribed to the Google Groups "Akka User List" group.
To post to this group, send email to akka...@googlegroups.com.
To unsubscribe from this group, send email to akka-user+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/akka-user?hl=en.
@Everyone else. So what's the model here? My thoughts are to have one or more MINA servers that accept client connections (or would Netty be better for this?) and then allocate a remote actor per client. I'm not sure if or why this would have to be client-initiated or server-initiated.
Basically the game client and the MINA server would speak a custom game protocol, then the MINA server would translate these game messages to messages to send to the remote actor. The point would be to take advantage of Akka's supervisor hierarchy for failure handling as well as clustering for wide scalability. Is this a reasonable approach?
Best,
Robert Boyd
Twitter: @rboyd
Thanks for your interest in Akka.
On 19 March 2010 19:16, Bobby Boyd <rb...@telematter.com> wrote:
>
> @Christopher did you get anywhere with this? I'm getting ready to try to do the same thing.
>
> @Everyone else. So what's the model here? My thoughts are to have one or more MINA servers that accept client connections (or would Netty be better for this?) and then allocate a remote actor per client. I'm not sure if or why this would have to be client-initiated or server-initiated.
The simplest approach would be to create one client-initiated actor
per client. The other way would be to have one
server-initiated/managed remote actor that receives a request to
create a "local" client actor (on the server), e.g. the remote actor
works as a broker.
>
> Basically the game client and the MINA server would speak a custom game protocol, then the MINA server would translate these game messages to messages to send to the remote actor. The point would be to take advantage of Akka's supervisor hierarchy for failure handling as well as clustering for wide scalability. Is this a reasonable approach?
That sounds good to me.
If you need remote supervision then you have to use client-initiated
remote actors.
Hope it helps, Jonas.
Hi Viktor.
Thanks for looking into it. I will try to reproduce again when I get into the office. Did you use the latest git pull on master, 'sbt console' ing in from the top directory?
This should of course work. It has worked fine before. I even have had attendees having lab code talking with their neighbours laptop on my training sessions.
Could be a regression.
I'll look into it tomorrow.
On 21 Apr 2010 17:53, "Ross McDonald" <ross...@gmail.com> wrote:
Hi Guys.
Sorry for the thread bump, just wondering has anyone actually run this on two different physical machines ? If so, could you walk me through the steps?
Thanks for your help, this is still a problem for me, not critical yet, but eventually I will need to spread the architecture out beyond one physical machine.
Cheers,
-- Ross
On 20 Apr 2010, at 10:09, Jonas Bonér wrote:
> Odd. Does it work if you type in the IP instead?
>
> On 20 April 2010 11:05, Ross McDonald <rossa...
> --
> You received this message because you are subscribed to the Google Groups "Akka User List" group.
...
Sorry for the delay on this. Lots of meetings and launches, all Akka
based you will be pleased to hear. I can confirm this does work as
expected, the problem was all my fault, too many versions of Akka
lying around for different work projects requiring different versions
of Scala. I did not pay enough attention to detail. I screwed up
with AKKA_HOME.
Sorry for wasting your time.