Unable to start lagom service in Kubernetes Cluster


(Gervais Siewe) #1

Hi,

I followed the example of the cipher project (https://github.com/lagom/lagom-java-chirper-example) in order to deploy my lagom services to a Kubernetes Cluster.
I’m using lagom version 1.4.1 and lagom14-scala-service-locator-dns version 2.3.0 for service discovery.
I created the Kubernetes template similar to the example in the cipher project. My service is able to connect to Cassandra, since in the cassandra logs, i can see that some tables for the keyspace of the service where created.
The problem is that when i deploy a statefulset of my service i always get the following error in the logs of the Kubernetes pod:

"[error] a.c.s.PersistentShardCoordinator - Persistence failure when replaying events for persistenceId [/sharding/UserStateEntityCoordinator]. Last known sequence number [0]

akka.pattern.CircuitBreaker$$anon$1: Circuit Breaker Timed out.

[error] a.c.s.PersistentShardCoordinator - Persistence failure when replaying events for persistenceId [/sharding/ChatRoomEventProcessorCoordinator]. Last known sequence number [0]

akka.pattern.CircuitBreaker$$anon$1: Circuit Breaker Timed out.

[error] a.c.s.PersistentShardCoordinator - Persistence failure when replaying events for persistenceId [/sharding/ChatRoomEntityCoordinator]. Last known sequence number [0]

akka.pattern.CircuitBreaker$$anon$1: Circuit Breaker Timed out."

I made some google search already, but i did find anything interesting.
Any Ideas what the problem here could be?


(Ignasi Marimon-Clos) #2

Hi Gervais,

the deployment process for kubernetes is being improved quite a bit. Also, because the chirper repository was getting a bit too convoluted we’ve split the repos into more specific equivalents.
We’re close to merging a new set of instructions on how to deploy chirper into kubernetes using the new Lightbend Orchsetration tools for Kubernetes. Here’s the PR:

Cheers,