PCS Cluster with Shared IP on different interface
I am creating a Fedora PCS Cluster for HAProxy. I have it running on VMWare, and am following this guide, and get to this step of adding a IPAddr2 resource: http://clusterlabs.org/doc/en-US/Pacemaker/1.1/html/Clusters_from_Scratch/_adding_a_resource.html
The only difference is that I need my cluster heartbeat/comms on one NIC/subnet, and my shared resource IP on a different NIC/subnet.
My internal comms is Node1=192.160.0.1 and Node2=192.168.0.2, and my resource ip is 10.0.0.1
How do I use this command in this situation:
pcs resource create ClusterIP ocf:heartbeat:IPaddr2 \
ip=192.168.0.120 cidr_netmask=32 op monitor interval=30s
If I add it as above, I get this:
[root@node-01 .pcs]# pcs status
Cluster name: mycluster
Last updated: Tue Oct 28 09:10:13 2014
Last change: Tue Oct 28 09:00:13 2014 via cibadmin on node-02
Stack: corosync
Current DC: node-02 (2) - partition with quorum
Version: 1.1.11-1.fc20-9d39a6b
2 Nodes configured
1 Resources configured
Online: [ node-01 node-02 ]
Full list of resources:
ClusterIP (ocf::heartbeat:IPaddr2): Stopped
Failed actions:
ClusterIP_start_0 on node-01 'unknown error' (1): call=7, status=complete, last-rc-change='Tue Oct 28 09:00:13 2014', queued=0ms, exec=27ms
ClusterIP_start_0 on node-02 'unknown error' (1): call=6, status=complete, last-rc-change='Tue Oct 28 09:00:13 2014', queued=0ms, exec=27ms
Answer
First, you need to specify the network device as mentioned by Daniel, e.g.
pcs resource create ClusterIP ocf:heartbeat:IPaddr2 ip=10.0.0.1 cidr_netmask=32 nic=eth0 op monitor interval=30s
Since you are running two node cluster, you don't have a fencing device. So you have to disable the STONITH setting but it is not recommended for production environment.
pcs property set stonith-enabled=false
The virtual ip address should be activated automatically.
#pcs status resources
Full list of resources:
ClusterIP (ocf::heartbeat:IPaddr2): Started:node-01