[Linux-cluster] Regd: Ethernet Channel Bonding issue in Cluster

Balaji balajisundar at midascomm.com
Wed Dec 26 10:25:18 UTC 2007


Dear All,
I have configured Cluster Suite with 2 servers
Server 1 : 192.168.13.110 IP Address and hostname is primary
Server 2 : 192.168.13.179 IP Address and hostname is secondary
Floating : 192.168.13.83 IP Address (Assumed by currently active server)

I have configured Ethernet Channel Bonding in Each Cluster Nodes and 
Channel Bonding Configuration Details are
1) Created bonding devices in "/etc/modprobe.conf" file 
    alias bond0 bonding
    options bonding miimon=100 mode=1
2) Edit the "/etc/sysconfig/network-scripts/ifcfg-eth0 and ifcfg-eth1" 
configuration
    DEVICE=eth0
    USERCTL= no
    ONBOOT=yes
    MASTER=bond0
    SLAVE=yes
    BOOTPROTO=none

    DEVICE=eth1
    USERCTL= no
    ONBOOT=yes
    MASTER=bond0
    SLAVE=yes
    BOOTPROTO=none
3) Created a network script for the bonding device is 
"/etc/sysconfig/network-scripts/ifcfg-bond0"
    DEVICE=bond0
    USERCTL=no
    ONBOOT=yes  
    NETMASK=255.255.255.0
    GATEWAY=192.168.13.1
    IPADDR=192.168.13.110
4) Reboot the system for the changes to take effect.

After i am rebooted both the server then cluster node becomes simplex 
and Services are started on both the nodes
The cluster ouput in primary node

Member Status: Quorate

  Member Name                              Status
  ----------- -------                              ---------
  primary                                         Online, Local, rgmanager
  secondary                                      Offline

  Service Name         Owner (Last)                   State
  --------- --------         -----------------                   --------
  Service                      primary                        started

The cluster ouput in secondary node

Member Status: Quorate

  Member Name                              Status
  ----------- -------                              ---------
  primary                                         Offline
  secondary                                      Online, Local, rgmanager

  Service Name         Owner (Last)                   State
  --------- --------         -----------------                   --------
  Service                      secondary                     started

Before Ethernet Channel Bonding cluster services are active in one node 
and other nodes acts as passive node.
But after Ethernet Channel Bonding cluster services are active on both 
the nodes
I need to solve the following issue and services are run only active node.

Can some one throw light on this peculiar problem

Regards
-S.Balaji




More information about the Linux-cluster mailing list