On my way to use Akka.NET for a scalable application, I am trying to setup a cluster of Lighthouse seed nodes. I am testing 3 Lighthouse nodes as seed nodes, each running on the same machine with different ports. Following is my hocon config sample:
lighthouse.actorsystem: "my-system"
# See petabridge.cmd configuration options here: https://cmd.petabridge.com/articles/install/host-configuration.html
petabridge.cmd.host = "0.0.0.0"
petabridge.cmd.port = 9111/9112/9113 #one in each node
akka.actor.provider = cluster
akka.remote.log-remote-lifecycle-events = DEBUG
akka.remote.dot-netty.tcp.transport-class = "Akka.Remote.Transport.DotNetty.TcpTransport, Akka.Remote"
akka.remote.dot-netty.tcp.applied-adapters = []
akka.remote.dot-netty.tcp.transport-protocol = tcp
akka.remote.dot-netty.tcp.public-hostname = "localhost"
akka.remote.dot-netty.tcp.hostname = "localhost"
akka.remote.dot-netty.tcp.port = 4001/4002/4003
akk.cluster.seed-nodes = ["akka.tcp://my-system@localhost:4001","akka.tcp://my-system@localhost:4002","akka.tcp://my-system@localhost:4003"]
akk.cluster.roles = [lighthouse]
If I start up these nodes from 3 command prompts, each is printing the following messages:
[INFO][22-01-2019 11:45:17][Thread 0020][Cluster] Cluster Node [akka.tcp://my-system@localhost:4001/4002/4003] - Node [akka.tcp://my-system@localhost:4001/4002/4003] is JOINING itself (with roles []) and forming a new cluster
[INFO][22-01-2019 11:45:17][Thread 0020][Cluster] Cluster Node [akka.tcp://my-system@localhost:4001/4002/4003] - Leader is moving node [akka.tcp://my-system@localhost:4001/4002/4003] to [Up]
My concern here is that, as per the logs printed, these three instances are not forming a cluster and seems to be forming three separate clusters as the nodes themselves are not getting any message about other Lighthouse nodes.
Can somebody please clarify if this is the expected behavior as there is no example seems to be available online.