[Pacemaker] [Problem]It does not start by single node configuration.

[Andrew Beekhof]

2010/1/28  <renayama19661014 at ybb.ne.jp>:
> Hi,
>
> I constituted a cluster in a single node by the next combination.
>
> * corosync 1.2.0
> * Pacemaker-1-0-2eed906f43e9(1.0.7)
> * Cluster-Resource-Agents-1545e3c22752
> * Reusable-Cluster-Components-1fa6e83c754c
>
> It started in a single node without arranging cib.xml.
>
> But, the node does not become the S_IDLE state.
> And the node does not become the online state.
>
> In the development version(Pacemaker-1-0-9287969750e), the same problem occurs.
> There was not a problem with version 1.0.6.
>
> Is setting necessary in the new version to start in a single node?
> Or is it a bug?

Very odd. CTS (successfully) tests this condition all the time.

What does corosync.conf look like?  Is there a firewall involved?
It _looks_ like the crmd isn't getting its own messages back.

Could you possibly retest with debug on?