Re: [akka-user] Re: Akka Microservice platform

2014-09-29 Thread Ryan Tanner
We're in the process of moving our Akka cluster over to CoreOS using etcd + skydns for service discovery. Our dev environment has been running on that for about two weeks now. Akka nodes publish their network location into etcd and then get their seed nodes out of the same, with skydns

Re: [akka-user] Re: Akka Microservice platform

2014-09-29 Thread Evan Chan
Ryan, That sounds really interesting. How are you doing leader election, if I may ask? Do all nodes register with etcd, or just the seed nodes? Is there some duplication of state and conflicts, ie Akka detects a node is down but the downed node still registers itself with etcd? On Mon, Sep

Re: [akka-user] Re: Akka Microservice platform

2014-09-29 Thread adriano santos
Hello guys, I see that the service discovery may closely resemble with the same approach used by DOSGI. http://cxf.apache.org/dosgi-architecture.html An approach seems to me easier adaptation to AKKA. Regards, Adriano Santos 2014-09-29 18:20 GMT-03:00 Ryan Tanner ryan.tan...@gmail.com: At

[akka-user] Re: Akka Microservice platform

2014-09-26 Thread Matlik
Has there been any movement forward on this discussion? I like the idea of having a single layer of cluster coordination (discovery and availability) for all Akka and non-Akka services. If there is a clear technical direction and code base, I'd be interested in possibly contributing.

[akka-user] Re: Akka Microservice platform

2014-06-16 Thread chris
On Monday, June 2, 2014 8:51:06 AM UTC-7, Evan Chan wrote: Hey guys, I would like to run an idea by the fine Akka community - which is to discuss what it would take to turn Akka into a platform for building a network of microservices: each one independently redeployable and easy to

[akka-user] Re: Akka Microservice platform

2014-06-11 Thread Vaughn Vernon
For the service discovery part, it might help to consider the old Jini model. I was going to work on this some time back, but ... Not that other ideas here aren't valuable, but I think that Jini got a lot of things right. Unfortunately JEE caught the limelight. Vaughn On Monday, June 2, 2014

[akka-user] Re: Akka Microservice platform

2014-06-09 Thread Stefan Podkowinski
Hi Let me add some notes based on my experience and further thoughts on this. Am Montag, 2. Juni 2014 17:51:06 UTC+2 schrieb Evan Chan: Here are some characteristics of such a platform: - Service discovery We've been using Zookeeper for discovering individual Akka based

[akka-user] Re: Akka Microservice platform

2014-06-09 Thread Todd Nist
Hi, Here is an alternative approach for service discovery with out using Zookeeper and has better synergy with AKKA clustering; Consul[0]. Consul is based on the Raft protocol[1] for consensus. But Consul also makes uses of Serf’s[2] gossip protocol for LAN and WAN membership status and

[akka-user] Re: Akka Microservice platform

2014-06-06 Thread Evan Chan
Hey Thomas, Sorry for the slow reply, for some reason I'm not getting email updates. RESTful endpoints: Yes, it is true these handle the polyglot thing very well, assuming standard HTTP and JSON.However, purely REST based services has many downsides: - only request/response is supported

[akka-user] Re: Akka Microservice platform

2014-06-02 Thread Thomas Lockney
Hey Evan, I'll bite... I guess my first question would be to ask for a more clear definition of what you mean by a microservice platform. It feels like this is trying to tackle too many different pieces of the ecosystem when some are already well-solved, while others just might not be