[Linux-cluster] cluster.conf not being propagated to all nodes

Digimer lists at alteeve.ca
Tue Dec 24 18:31:04 UTC 2013

On 24/12/13 01:21 PM, Neale Ferguson wrote:
> What I'm really after is confirmation that it's luci which is propagating the cluster.conf via the ricci service running on the nodes. If that's the case, I want to determine why luci only opts to send to one node and not the other.

Log into a node, manually edit cluster.conf to increase 
config_version="x", save and exit, run 'ccs_config_validate' and if 
there are no errors, run 'cman_tool version -r'. Enter the 'ricci' 
password(s) if prompted. Then on both nodes run 'cman_tool version'. If 
the returned version is the new one you set, then the config pushed out 
successfully and it would indicate that luci is the problem.

Please run 'tail -f -n 0 /var/log/messages' on both nodes prior to 
'cman_tool version -r'. If the config fails to push out, please copy the 
log output from both nodes to here.


Papers and Projects: https://alteeve.ca/w/
What if the cure for cancer is trapped in the mind of a person without 
access to education?

More information about the Linux-cluster mailing list