Hi All,
This is the log for the case. The node 10.1.4.196 is removed at 14:08 due to machine reboot, the client keeps trying to connect to the elasticsearch cluster but fails. Master Node : [2014-03-08 14:08:26,531][INFO ][cluster.service ] [10.1.4.197:9202] removed {[10.1.4.196:9202][_sJrum34QWGqEkv8CvAtow][inet[/10.1.4.196:9302]],}, reason: zen-disco-node_failed([10.1.4.196:9202][_sJrum34QWGqEkv8CvAtow][inet[/10.1.4.196:9302]]), reason failed to ping, tried [3] times, each with maximum [30s] timeout Client : 2014-03-08 14:15:36,184 WARN org.elasticsearch.transport.netty - [Bulldozer] exception caught on transport layer [[id: 0x50dc218f]], closing connection java.net.NoRouteToHostException: No route to host (The cluster health at this moment is Yellow and there is no unassigned shard.) The node is back at 14:25, the client can successfully connected to the cluster again. Client : 2014-03-08 14:25:20,597 WARN org.elasticsearch.transport.netty - [Bulldozer] exception caught on transport layer [[id: 0xf24d85d7]], closing connection java.net.NoRouteToHostException: No route to host Master Node : [2014-03-08 14:25:57,984][INFO ][cluster.service ] [10.1.4.197:9202] added {[10.1.4.196:9202][rFZ7k7XSSY231EgPoDfmFw][inet[/10.1.4.196:9302]],}, reason: zen-disco-receive(join from node[[10.1.4.196:9202][rFZ7k7XSSY231EgPoDfmFw][inet[/10.1.4.196:9302]]]) (The cluster health at this moment is Green.) In the above case, the client should be able to connect to the cluster even a node is removed from the cluster. For the client, the connection is created as followings : Settings settings = ImmutableSettings.settingsBuilder() .put("cluster.name", "clustername") .put("client.transport.sniff", true) .build(); TransportClient client = new TransportClient(settings); client.addTransportAddress(new InetSocketTransportAddress( "10.1.4.195" /* hostname */, 9300 /* port */)); client.addTransportAddress(new InetSocketTransportAddress( "10.1.4.196" /* hostname */, 9300 /* port */)); client.addTransportAddress(new InetSocketTransportAddress( "10.1.4.197" /* hostname */, 9300 /* port */)); The master node is 10.1.4.197 while the node being removed is 10.1.4.196. For the cluster setting, all setting is using the default except the the discovery.zen.minimum_master_nodes which is set to 3. Is there any problem for the above setting which cause this issue? Thanks. -- You received this message because you are subscribed to the Google Groups "elasticsearch" group. To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/b1f3adf5-723b-49aa-bffe-674c5ce930e5%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.