[Pacemaker] corosynce cluster - new - 2nd won't join first

Phil Hunt Phil.Hunt at orionhealth.com
Wed Nov 2 14:19:53 UTC 2011


Hi

I've put together 4 clusters so far, they have worked fine.

I've been tasked to put together a 5th, but the second node won't join the
first!

The only difference is that the netmask for all the others was
255.255.255.0

The netmask assigned to this one is 255.255.255.224

The machine addresses are 10.115.143.13 and .14.  They ping fine

So in the corosync.conf, I put in the bcast address shown on the servers:
eth0      Link encap:Ethernet  HWaddr 00:50:56:89:00:19  
          inet addr:10.115.143.14  Bcast:10.115.143.31
Mask:255.255.255.224

In corosync.conf:
        interface {
                ringnumber: 0
 
                # The following three values need to be set based on your
environment
                #bindnetaddr: 10.115.143.0
                bindnetaddr: 10.115.143.31
                mcastaddr: 226.94.1.1
                mcastport: 4001
        }

Corosync installed on both without problem, RHEL 5.7.  I have no services
running yet, just testing
The two connecting.  The first node shows:
============
Last updated: Wed Nov  2 10:16:29 2011
Stack: openais
Current DC: nchmartstmsg02 - partition WITHOUT quorum
Version: 1.0.11-1554a83db0d3c3e546cfd3aaff6af1184f79ee87
1 Nodes configured, 2 expected votes
0 Resources configured.
============

Online: [ nchmartstmsg02 ]

In the logs on nchmartstmsg01, I see NOTHING about 02 trying to connect.
It is like they cannot talk together.
I do not understand the mcastaddr.  Does that have to change for a
different subnet mask too?

Thanks in advance


Phil Hunt 
Orion Health
Senior AMS Consultant 

DID: +1 857 488 4749
Cell:  +1 508 654 7371
Skype: philhu0724 

 










More information about the Pacemaker mailing list