Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • Apr 04 2018 06:06
    ktoso commented #179
  • Apr 04 2018 06:05
    ktoso commented #179
  • Apr 03 2018 08:31
    hseeberger opened #179
  • Apr 03 2018 06:50
    franbh commented #174
  • Apr 03 2018 06:39
    ktoso commented #174
  • Apr 03 2018 06:32
    markusjura commented #174
  • Apr 03 2018 02:56
    ktoso commented #174
  • Apr 03 2018 02:54
    ktoso commented #174
  • Mar 12 2018 09:44
    hseeberger commented #173
  • Mar 12 2018 09:41

    hseeberger on v0.19.0

    (compare)

  • Mar 12 2018 09:41

    hseeberger on master

    Upgrade to sbt 1.1.1 Prepare release 0.19.0 (compare)

  • Mar 12 2018 09:31
    hseeberger commented #173
  • Mar 12 2018 09:31

    hseeberger on master

    Restart FSM process on seed nod… (compare)

  • Mar 12 2018 09:31
    hseeberger closed #168
  • Mar 12 2018 09:31
    hseeberger closed #173
  • Mar 11 2018 16:54
    nick-nachos commented #173
  • Mar 11 2018 11:39
    nick-nachos commented #173
  • Mar 11 2018 11:30
    nick-nachos synchronize #173
  • Mar 10 2018 06:53
    hseeberger commented #173
  • Mar 09 2018 22:49
    dmsquires commented #173
Cheng Wei
@weicheng113
Thanks
We are using AWS Container Service docker to build the cluster and Managed Redis service as coordination service.
Heiko Seeberger
@hseeberger
@weicheng113 that's probably an issue with the Redis implementation of the coordination service which is a community project – please ask there.
Juan José Vázquez
@juanjovazquez
Hi, @hseeberger. Just to know, do you have plans to publish to ConstructR 0.16 to Maven Central any time soon?. Thanks!. /cc @gerson24
Heiko Seeberger
@hseeberger
@juanjovazquez ah, sure, thanks for the reminder. Damn manual syncing :-(
Juan José Vázquez
@juanjovazquez
@hseeberger yeah, I know, it's a real pain. Maybe the Scala Center could help in the future promoting a better and easier ecosystem of tools.
Heiko Seeberger
@hseeberger
Currently syncing 0.16.1 ...
Ievgenii Shepeliuk
@eshepelyuk
Hi, all
I'm curious if there any example of using AKKA ClusterClient with ConstructR ?
Heiko Seeberger
@hseeberger
Not that I'm aware of. How are they related?
Ievgenii Shepeliuk
@eshepelyuk
Hmmm, i want to access cluster bootstrapped by ConstructR using ClusterClient
This requires knowing seed nodes, but in ConstructR case, they can be only obtained from coordination software (zookeeper, consul)
Ievgenii Shepeliuk
@eshepelyuk
The only way to obtian I see now - is to use low level API of the corresponding coordination provider.
maybe there's better way ?
Heiko Seeberger
@hseeberger
Interesting!
No better way yet.
Ievgenii Shepeliuk
@eshepelyuk
I'm not sure yet about our setup. Maybe we have to join all our service into the single cluster.
Heiko Seeberger
@hseeberger
Not sure if it's a good idea to even make the coordination service accessible from the outside, though!
Ievgenii Shepeliuk
@eshepelyuk
Actually our clustering needs are boiling down to sharding.
Heiko Seeberger
@hseeberger
You could write a HTTP endpoint which also is a cluster member listener and expose nodes that way.
Ievgenii Shepeliuk
@eshepelyuk
Or we can join all our services into the single AKKA cluster.
Ievgenii Shepeliuk
@eshepelyuk
What do you think about creating dedicated API to obtain ClusterClient using the same coordination nodes as for cluster node itself ?
So one can implement corresponding low level calls to zookeeper, etcd
I mean, I can try to create a PR if you find this idea aproppriate :)
Heiko Seeberger
@hseeberger
Not sure what you mean exactly.
Ievgenii Shepeliuk
@eshepelyuk
Ok. nevermind. thanx
LMalikov
@LMalikov
Hello, I'm trying to use ConstructR in the Kubernetes world and I encountered a problem with Pod re-creation
after the main Akka cluster node is brought down and then up again (with new IP), it seems cluster is unable to form properly again as all the nodes try to join old leader node IP (which is already invalid)
even when key for old address expires in etcd, every node still tries to join it
so am I missing something or is ConstructR not a suitable solution for K8s paradigm?
or maybe it is a known issue and there is a known workaround? (:
any help would be much appreciated, thanks
LMalikov
@LMalikov
anyone, please?
Heiko Seeberger
@hseeberger
I'm not sure I understand. What do you mean with "main Akka cluster node"? In Akka Cluster there is no such thing like a master. Also, if you only down one node why should the others try to join? They are still members of the (now smaller) cluster.
Salar Rahmanian
@softinio
I am trying to use constructr to deploy akka using kubernetes
do I need to use a specific version of etcd with constructr?
Heiko Seeberger
@hseeberger
Yes, currently etcd 2.
Salar Rahmanian
@softinio
@hseeberger thx
Abbass Marouni
@amarouni
Hello guys,
I facing a weird issue with Akka cluster & constructr-zookeeper. I'm running the cluster in Docker containers, everything works as expected but when I restart a container it cannot re-join the cluster. The cluster keeps trying to reach the node on its old port (I'm using random ports to) which is not reachable anymore.
Is this something the you already saw with Akka cluster & constructr or this is purely Akka cluster
Or I'm missing something here.
Fran Bermejo
@franbh
Hi @hseeberger, it seems that ConstructR 0.18.1 for Scala 2.11 is missing from Maven Central
Heiko Seeberger
@hseeberger
Yes, it is missing.
MohamedShawky85
@MohamedShawky85
Hi Guys i have an issue now with akka cluster joining
After restarting the running cluster nodes with another ones it fails and returns to me an exception " Terminating the system, because constructr-machine has terminated!'
[ERROR] [12/13/2017 11:47:10.585] [MI_Cluster_Docker-akka.actor.default-dispatcher-18] [akka.tcp://MI_Cluster_Docker@10.131.2.148:2552/system/constructr/constructr-machine] Timeout in Joining!
[ERROR] [12/13/2017 11:47:10.587] [MI_Cluster_Docker-akka.actor.default-dispatcher-20] [akka.tcp://MI_Cluster_Docker@10.131.2.148:2552/system/constructr] Terminating the system, because constructr-machine has terminated!
@hseeberger Any Help please
Abbass Marouni
@amarouni
@MohamedShawky85 Does it look like this issue : hseeberger/constructr#168
Dmitriy Zakomirnyi
@dmi3zkm
Hi all!
I guess it's a lame question, but let me ask anyway :smile:
What if I use constructr with akka cluster autodown feature.
Can I have a split brained cluster as a result? Doesn't zookeeper supposed to coordinate nodes to avoid it?
Ievgenii Shepeliuk
@eshepelyuk
BTW, we're using ConstructR in Docker Swarm environment
Ievgenii Shepeliuk
@eshepelyuk
@/all plz anyone ?
Fabrice Sznajderman
@Fabszn
Hi all
@/all Is there a way to ensure that the cluster in correctly formed?
Ievgenii Shepeliuk
@eshepelyuk
@Fabszn this chat seems to be dead
Fabrice Sznajderman
@Fabszn
@eshepelyuk yes :(
Heiko Seeberger
@hseeberger
The project has been archived.
charles adetiloye
@charlesa101
Any reason for archiving the project @hseeberger ?
Heiko Seeberger
@hseeberger
Yes, please take a look at the respective issue.