Take the 2-minute tour ×
Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

I'm having some troubles when configuring pacemaker+corosync between Centos 5 and Centos 6. Here is ouput of crm_mon:

On node1:

Last updated: Sun Jul 21 19:02:21 2013
Last change: Sun Jul 21 18:14:48 2013 via crmd on svr077-53149.localdomain
Stack: openais
Current DC: svr077-53149.localdomain - partition WITHOUT quorum
Version: 1.1.8-2.el5-394e906
2 Nodes configured, 2 expected votes
1 Resources configured.


Online: [ svr077-53149.localdomain ]
OFFLINE: [ svr423L-2737.localdomain ]

Crond   (lsb:crond):    Started svr077-53149.localdomain

On node2:

Last updated: Sun Jul 21 19:03:40 2013
Last change: Sun Jul 21 18:14:56 2013
Stack: classic openais (with plugin)
Current DC: NONE
1 Nodes configured, 2 expected votes
0 Resources configured.


ONLINE: [ svr423L-2737.localdomain ]

Here is my corosync log

My question is:

  • Why each node has their own DC, node1 detects two nodes while node2 show only one?
  • What can be problem make two node can't join cluster?
  • Can we make pacemaker+corosync between Centos 5 and Centos 6?

Here is my software version in two node:

Node1:

 - Corosync version 1.4.3 
 - Pacemaker version 1.1.8-2.el5 
 - Centos release 5.8 (Final)

And

Node2:
 - Corosync version 1.4.1
 - Pacemaker version 1.1.8-7.el6
 - Centos release 6.4 (Final)

UPDATE When I configured the first time, everything worked ok. After I shutdown node 1, turn on it to test failover case, this problem occured.

share|improve this question
    
what about settings of firewall/selinux on each of the nodes? –  ALex_hha Jul 21 '13 at 12:55
    
@ALex_hha: they are disabled. You can see in log file that the communication between two node is ok. –  cuonglm Jul 21 '13 at 14:12

1 Answer 1

If you are using multicast, then check your igmp support in your switch and in your cluster hosts check the corosync membershit status

corosync-cfgtool -s corosync-cmapctl | grep mem

Thanks

share|improve this answer
    
The first time I config, everything works fine. When I shutdown one node to test failover case, the problem occurs. –  cuonglm Jul 28 '13 at 6:12

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.