Description of problem:
Bad node(s) in error message(s) when executing maintenance/unmaintenance commands with more nodes which aren't in cluster.

Version-Release number of selected component (if applicable):
pcs-0.9.148-7.el6.x86_64

How reproducible:
always

Steps to Reproduce:

Cluster nodes:
[root@virt-242 ~]# pcs status nodes config
Corosync Nodes:
 virt-242 virt-243 virt-262
Pacemaker Nodes:
 virt-242 virt-243 virt-262

1. Try these commands:

a)
[root@virt-242 ~]# pcs node  maintenance virt-00{1..3}
Error: Node 'virt-001' does not appear to exist in configuration
Error: Node 'virt-001' does not appear to exist in configuration
Error: Node 'virt-001' does not appear to exist in configuration

b)
[root@virt-242 ~]# pcs node  unmaintenance virt-00{1..3}
Error: Node 'virt-001' does not appear to exist in configuration
Error: Node 'virt-001' does not appear to exist in configuration
Error: Node 'virt-001' does not appear to exist in configuration

c)
[root@virt-242 ~]# pcs node  maintenance virt-242 virt-00{1..3}
Error: Node 'virt-242' does not appear to exist in configuration
Error: Node 'virt-242' does not appear to exist in configuration
Error: Node 'virt-242' does not appear to exist in configuration
[root@virt-242 ~]# echo $?
1
[root@virt-242 ~]# pcs cluster cib | grep maintenance
          <nvpair id="nodes-virt-242-maintenance" name="maintenance" value="on"/>

NOTE: The maintenance mode was set despite the exit error code.

Actual results:
Bad node in error messages. First given node is always in each error message.

Expected results:
Right node names in error messages.

Additional info:
If cluster node and other hosts are mixed, maintenance mode is set on the cluster nodes and error messages are written for others. Command exits with error code, which is inconsistent.