Description of problem:
I had a three node cluster up (taft-0[123]) and when I attempted to add taft-04
to it, I saw the following error:

"An unknown ricci error occurred on taft-02.lab.msp.redhat.com:11111"

I don't see a cluster.conf file on taft-04 and openais doesn't list it's ip
address (10.15.89.70) as a member where luci is running:

Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ] CLM CONFIGURATION CHANGE
Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ] New Configuration:
Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ] Members Left:
Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ] Members Joined:
Jul 23 14:51:42 taft-01 openais[9026]: [SYNC ] This node is within the primary
component and will provide service.
Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ] CLM CONFIGURATION CHANGE
Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ] New Configuration:
Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ]  r(0) ip(10.15.89.67)
Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ]  r(0) ip(10.15.89.68)
Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ]  r(0) ip(10.15.89.69)
Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ] Members Left:
Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ] Members Joined:
Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ]  r(0) ip(10.15.89.67)
Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ]  r(0) ip(10.15.89.68)
Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ]  r(0) ip(10.15.89.69)
Jul 23 14:51:42 taft-01 openais[9026]: [SYNC ] This node is within the primary
component and will provide service.
Jul 23 14:51:42 taft-01 openais[9026]: [TOTEM] entering OPERATIONAL state.
Jul 23 14:51:42 taft-01 openais[9026]: [CMAN ] quorum regained, resuming activity
Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ] got nodejoin message 10.15.89.67
Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ] got nodejoin message 10.15.89.68
Jul 23 14:51:42 taft-01 openais[9026]: [CLM  ] got nodejoin message 10.15.89.69
Jul 23 14:51:42 taft-01 openais[9026]: [CPG  ] got joinlist message from node 1
Jul 23 14:51:42 taft-01 openais[9026]: [CPG  ] got joinlist message from node 4
Jul 23 14:51:43 taft-01 ccsd[9020]: Initial status:: Quorate
Jul 23 14:52:45 taft-01 ccsd[9020]: Update of cluster.conf complete (version 8
-> 9).
Jul 23 14:52:52 taft-01 luci[6716]: Error reading from
taft-02.lab.msp.redhat.com:11111: timeout


[root@taft-01 ~]# cman_tool nodes
Node  Sts   Inc   Joined               Name
   1   M    192   2007-07-23 14:51:42  taft-02.lab.msp.redhat.com
   2   M    192   2007-07-23 14:51:42  taft-01.lab.msp.redhat.com
   3   X      0                        taft-04.lab.msp.redhat.com
   4   M    192   2007-07-23 14:51:42  taft-03.lab.msp.redhat.com


Version-Release number of selected component (if applicable):
luci-0.10.0-2.el5
cman-2.0.69-1.el5

How reproducible:
I've seen this quite a few times now