Hi Mahmood,

it's somewhat strange: One one hand, CIB is reported as unreachable, but on the other hand you can query it. One one hand, "crm_mon -1" reports both nodes as "offline", but your CIB reports <node_state ... crmd="online" /> for both nodes. Did you fetch this cibadmin output when the node were in one of the failed states you reported earlier?

What about the expected cib messages in syslog, or the question whether that process is alive (when you see the failure) - did you get a chance to verify that? It somehow looks lib the cib process is failing, rendering pacemaker useless. If this can be confirmed, the question is "why".

This whole situation appears rather complicated and hard to be handled off-site via forums. If you have a support contract with SUSE, I'd recommend to open a service request, stating that you're evaluating HAE prior to purchase and have run into this problem - and if no support contract is available, see your SUSE / vendor pre-sales support for assistance.

I'll still try to help as best as I can, of course.