0
votes

I have 3 machines.

  1. 172.30.30.57 for Hazelcast management center
  2. 172.30.30.56 for service number 1 and
  3. 172.30.30.106 for service number 2

config files for these two services are:

<group>
        <name>dev</name>
    </group>
    <management-center enabled="true">http://172.30.30.57:8080/hazelcast-mancenter-3.12.4</management-center>


<join>
            <multicast enabled="false">
                <multicast-group>224.2.2.3</multicast-group>
                <multicast-port>54327</multicast-port>
            </multicast>
            <tcp-ip enabled="true">
                <interface>172.30.30.106</interface>
                <member-list>
                    <member>172.30.30.56:5701</member>
                   <member>172.30.30.106:5701</member>
                </member-list>
            </tcp-ip>

and

 <group>
        <name>dev</name>
    </group>
    <management-center enabled="true">http://172.30.30.57:8080/hazelcast-mancenter-3.12.4</management-center>


<join>
            <multicast enabled="false">
                <multicast-group>224.2.2.3</multicast-group>
                <multicast-port>54327</multicast-port>
            </multicast>
            <tcp-ip enabled="true">
                <interface>172.30.30.106</interface>
                <member-list>
                    <member>172.30.30.56:5701</member>
                   <member>172.30.30.106:5701</member>
                </member-list>
            </tcp-ip>

The problem is that Hazelcast management center just shows one of them each time and they do not see each other.

how should i change the configs to have a cluster of hazelcast grouped nodes?

2
Can you please share the member logs? The problem is probably not the Management Center. It's probably that the members are not forming a cluster. Also, you need <network> tag around <join> in your Hazelcast config (not sure if you've omitted it only here for brevity or it's like that in your application as well).emre
I changed the group names and management center shows both services in two distinct groups. so you are right, there is something wrong in clustering. the <network> tag is omitted in descriptions for brevity.Mehdi
this is the error log: Could not connect to: /172.30.30.56:5701. Reason: SocketException[No route to host to address /172.30.30.56:5701]Mehdi

2 Answers

0
votes

Both member's XMLs are you've posted have

<interface>172.30.30.106</interface>

Could that be your problem ?

0
votes

The problem solved. I followed the Exception: "No route to host to address...", and found that I can't connect to other addresses even by telnet. so I turned off the firewall and SELinux in centos and connection established.