[Pacemaker] Basic Clone IP configuration not starting automatically

[Net Warrior]

Hi there guys.
I'm new to pacemaker and I'm trying to accomplish an active/active cluster
with a VIP ip address, I used the java frontend for the initial setup, I'm
using LCMC-1.7.6.jar.


This is what I've got:
CentOS release 6.5
pacemaker-libs-1.1.10-14.el6_5.3.x86_64
pacemaker-cli-1.1.10-14.el6_5.3.x86_64
pacemaker-cluster-libs-1.1.10-14.el6_5.3.x86_64
pacemaker-1.1.10-14.el6_5.3.x86_64
pcs-0.9.90-2.el6.centos.3.noarch

Cluster configuration:

Last updated: Thu Feb  5 14:27:41 2015
Last change: Thu Feb  5 14:17:26 2015 via cibadmin on haproxy02
Stack: cman
Current DC: haproxy01 - partition with quorum
Version: 1.1.10-14.el6_5.3-368c726
2 Nodes configured
2 Resources configured


Online: [ haproxy01 haproxy02 ]

 Clone Set: cl_IPaddr2_1 [res_IPaddr2_virtualip] (unique)
     res_IPaddr2_virtualip:0    (ocf::heartbeat:IPaddr2):       Started
haproxy01
     res_IPaddr2_virtualip:1    (ocf::heartbeat:IPaddr2):       Started
haproxy01

Node Attributes:
* Node haproxy01:
* Node haproxy02:

Clone IP configuration:
pcs resource show cl_IPaddr2_1
 Clone: cl_IPaddr2_1
  Meta Attrs: clone-max=2 clone-node-max=2 notify=true globally-unique=true
  Resource: res_IPaddr2_virtualip (class=ocf provider=heartbeat
type=IPaddr2)
   Attributes: ip=1.1.1.1
   Meta Attrs: target-role=started
   Operations: stop interval=0 timeout=20 (op-res_IPaddr2_virtualip-stop)
               monitor interval=5s
(res_IPaddr2_virtualip-monitor-interval-5s)
               start interval=0 (res_IPaddr2_virtualip-start-interval-0)

IPTABLES
Chain INPUT (policy ACCEPT)
target     prot opt source               destination
CLUSTERIP  all  --  0.0.0.0/0            1.1.1.1       CLUSTERIP
hashmode=sourceip-sourceport clustermac=51:72:D8:57:74:84 total_nodes=2
local_node=1 hash_init=0

CMAN configuration:

<?xml version="1.0"?>
<cluster config_version="1" name="clustername">
   <logging debug="off"/>
   <clusternodes>
     <clusternode name="haproxy01" nodeid="1">
       <fence>
         <method name="pcmk-redirect">
           <device name="pcmk" port="haproxy01"/>
         </method>
       </fence>
     </clusternode>
     <clusternode name="haproxy02" nodeid="2">
       <fence>
         <method name="pcmk-redirect">
           <device name="pcmk" port="haproxy02"/>
         </method>
       </fence>
     </clusternode>
   </clusternodes>
   <fencedevices>
     <fencedevice name="pcmk" agent="fence_pcmk"/>
   </fencedevices>
</cluster>


If I do an sshd to the VIP I can see the round robin is woking and it
switches bettween hosts and I can see of course the alias created on the
eth0 iface.

My problem is that when I reboot a server to test it, the alias is not
created automatically and I have to perform a stop/start from the java
frontend.

What am I missing to make it work automatically?

Thanks for your time and support.
Regards
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20150205/aabcf044/attachment-0003.html>