Re: [jgroups-users] delay in receving are-you-alive message

[Bela B. <be...@ya...>]

Picking 224.0.0.x is not recommended, so I suggest pick a different 
mcast address.
However, that's not the problem.

I would pick 232.1.1.1 / 45566 for node1 and node2 and
232.2.2.2 / 56677 for nocde3 and node4

On 11/19/12 12:54 PM, Jasmine_11 wrote:
>
> In our application , we have 2 group ,each group has 2 nodes and all in one
> subnet (24.25.9.*) . We have used common multi cast ip (224.0.0.1) for 2
> groups and but we have used two different port for each group.
>
> PollerClusterGroup001   - (224.0.0.1 / 45556)  node1 -(24.25.9.139) ,node2
> -(24.25.9.140)
> PollerClusterGroup002    -(224.0.0.1/45557)   node 1  -(24.25.9.137 ),node2-
> 24.25.9.138)
>
> As said each group has 2 nodes . initially each nodes are communicating to
> the corresponding groups without any issue. after nodes are running in a day
> ,we are seeing the following message , one node left from the group .We
> checked with our system admin team that there is no network problem during
> the issue reproduced , CPU , Memory usage are very less .
>
> [14 Nov 2012 18:36:43:228] pbcast.GMS:node1: not member of view [node2|2];
> discarding it
>
>
> is there any problem using same mulitcate Ip (224.0.0.1) for 2 different
> groups? if all the four nodes in the same network (LAN)
>
> - the cluster split
> - The cluster didn't form in the first place (e.g. in correct config,
> firewall etc)
>
> ( We dont have any firewall issue between nodes , we don't have any problem
> while starting of the nodes .it was running for a day successfully, we see
> this issue after a day .
>
>
>
> Bela Ban wrote:
>>
>> This means that you got a message from a node that's not a current
>> member of the cluster. In JGroups, only cluster members can send
>> messages; messages from non-members are discarded.
>>
>> Why this is the cae, I can't tell (you provided no useful information),
>> but most likely:
>> - the cluster split
>> - The cluster didn't form in the first place (e.g. in correct config,
>> firewall etc)
>>
>> On 11/15/12 5:06 PM, Jasmine_11 wrote:
>>>
>>>
>>>
>>> As suggested by you we migrated to latest version of jgroups 3.2.0
>>> version.but still we are facing same issue. found below message in the
>>> logs
>>> directory.We are using UDP communication and used default  udp.xml file.
>>>
>>> [14 Nov 2012 18:36:43:228] pbcast.GMS:testmachine1-35870: not member of
>>> view
>>> [testamachine2-53653|2]; discarding it
>>>
>>> Due to above issue , both nodes in active/active state.
>>>
>>> Please let me know if you need any more details on this
>>>
>>>
>>> Bela Ban wrote:
>>>>
>>>>
>>>>
>>>> On 10/18/12 8:45 PM, Jasmine_11 wrote:
>>>>>
>>>>> Hi,
>>>>>
>>>>> Our Applications is running with jgroups 2.6.3.CR2.
>>>>
>>>>
>>>> I don't support 2.6.x, see https://community.jboss.org/wiki/Support.
>>>> Also, even on the 2.6 branch, 2.6.3 is way behind the latest (2.6.21)
>>>> release.
>>
>>
>> --
>> Bela Ban, JGroups lead (http://www.jgroups.org)
>>
>> ------------------------------------------------------------------------------
>> Monitor your physical, virtual and cloud infrastructure from a single
>> web console. Get in-depth insight into apps, servers, databases, vmware,
>> SAP, cloud infrastructure, etc. Download 30-day Free Trial.
>> Pricing starts from $795 for 25 servers or applications!
>> http://p.sf.net/sfu/zoho_dev2dev_nov
>> _______________________________________________
>> javagroups-users mailing list
>> jav...@li...
>> https://lists.sourceforge.net/lists/listinfo/javagroups-users
>>
>>
>

-- 
Bela Ban, JGroups lead (http://www.jgroups.org)