Did you open 6001 as well? The documentation states that 6001 is used to ' listen for cluster communications on the network interface'..
M On Monday, 1 September 2014 20:48:18 UTC+1, Kishore Kumar Garg wrote: > > Yes, I also tried with setting up ha.server=192.168.1.3:6001 but got the > same result. > Port 5001 is open on all the nodes. However, I am not sure if I have to > open any other port too for setting up the cluster. > > Please let me know the port numbers you open in your cluster. > > Thanks, > Kishore > > On Monday, 1 September 2014 23:25:25 UTC+5:30, Mark Findlater wrote: >> >> When running on a single server I guess you set all the ports >> differently, when running on multiple servers did you also set the >> ha.server value? >> >> I set allow_init_cluster to true as who knows which server I will start >> first on any given day! >> >> What messages are you seeing in the logs when starting up? >> >> The likelihood is that this is configuration of your network, i.e are you >> sure each node can see the other nodes on the ports required (I know you >> said you can telnet to each of them on 5001, can they equally contact each >> other?) >> >> My cluster isn't running at the moment, but I can start it up again >> tomorrow and remind myself which ports I had to open up in addition to 5001 >> (6001 will be in the list).. >> >> M >> >> >> On Monday, 1 September 2014 18:41:55 UTC+1, Kishore Kumar Garg wrote: >>> >>> Thanks Mark. >>> >>> However, even after giving the values of ha.cluster_server, I was not >>> able to start it. I tried the same setup on single server and I was able to >>> start 2 embedded server instances of Neo4j and one arbiter instance. >>> >>> Then, I was trying to replicate the same setup on 3 different hosts and >>> it starts breaking. I didn't set allow_init_cluster configuration value as >>> true because in that case Host 2 would have created a separate cluster >>> which would have defied the whole purpose. >>> >>> I am still clueless why it was working on same host but not on different >>> host which are in same domain. >>> >>> -Kishore >>> >>> On Monday, 1 September 2014 21:39:40 UTC+5:30, Mark Findlater wrote: >>>> >>>> If Host 1 is starting OK, does it have the allow_init_cluster >>>> configuration value set to true? You can verify that it is failing to join >>>> by setting the ha.cluster_join_timeout configuration entry to some value >>>> in >>>> seconds. >>>> >>>> I think each of your ha.initial_hosts values should contain all of the >>>> servers that you expect to see, although I would also expect this not to >>>> be >>>> a terminal as you ought to be able to add new servers into the cluster >>>> without reconfiguring each server! (Refering to the host the server is >>>> running on is AOK (from ha setup documentation >>>> <http://docs.neo4j.org/chunked/stable/ha-setup-tutorial.html>) - >>>> ha.initial_hosts is a comma separated list of address/port pairs, which >>>> specify how to reach other Neo4j instances in the cluster (as configured >>>> via their ha.cluster_server option). These hostname/ports will be used >>>> when >>>> the Neo4j instances starts, to allow it up to find and join the cluster. >>>> Specifying an instance’s own address is permitted.) >>>> >>>> You should set a value for ha.cluster_server otherwise you cannot be >>>> entirely sure what the embedded instance has bound to (ha.cluster_server= >>>> 192.168.1.3:5001 for server 1, etc). >>>> >>>> I had a few issues with this, some caused by Spring Data not being HA >>>> aware, and spent a bit of time looking at the ClusterJoin, ClusterClient >>>> and ClusterSettings classes. >>>> >>>> M >>>> >>>> >>>> >>>> On Monday, 1 September 2014 16:12:53 UTC+1, Kishore Kumar Garg wrote: >>>>> >>>>> Current behavior:- >>>>> >>>>> Host 1 starts perfectly fine. However, when I start the 2nd host it >>>>> hangs for life time and never came out of it. >>>>> I noticed this happens when it is not able to connect to the cluster >>>>> created by Host 1. >>>>> >>>>> Please let me know if I am missing something. >>>>> >>>>> Thanks, >>>>> Kishore >>>>> >>>>> On Monday, 1 September 2014 20:03:11 UTC+5:30, Kishore Kumar Garg >>>>> wrote: >>>>>> >>>>>> Hi, >>>>>> >>>>>> I am setting up a HA cluster of Neo4J embedded server(with spring big >>>>>> data) and Neo4j version is 2.0.3. >>>>>> However, it is not getting started though hosts are in same network >>>>>> and I am able to telnet on port 5001 on each node. >>>>>> >>>>>> Please find below the HA properties for each node below:- >>>>>> >>>>>> Host 1:- >>>>>> ------------ >>>>>> ha.server_id=1 >>>>>> ha.initial_hosts=192.168.1.3:5001 >>>>>> org.neo4j.server.database.mode=HA >>>>>> >>>>>> >>>>>> Host 2:- >>>>>> ------------ >>>>>> ha.server_id=2 >>>>>> ha.initial_hosts=192.168.1.3:5001,192.168.1.4:5001 >>>>>> org.neo4j.server.database.mode=HA >>>>>> >>>>>> >>>>>> Host 3:- >>>>>> ------------ >>>>>> ha.server_id=3 >>>>>> ha.initial_hosts=192.168.1.3:5001,192.168.1.4:5001,192.168.1.5:5001 >>>>>> org.neo4j.server.database.mode=HA >>>>>> >>>>>> This configuration is not working. Please let me know what is missing >>>>>> here. >>>>>> Any help is highly appreciated. >>>>>> >>>>>> Thanks in advance, >>>>>> Kishore >>>>>> >>>>> -- You received this message because you are subscribed to the Google Groups "Neo4j" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
