Automation with Power

Power Business Continuity and Automation

Connect, learn, and share your experiences using the business continuity and automation technologies and practices designed to ensure uninterrupted operations and rapid recovery for workloads running on IBM Power systems. 


#Power
#TechXchangeConferenceLab

 View Only
  • 1.  HACMP Questions

    Posted Wed January 20, 2010 01:00 AM

    Originally posted by: GurpreetSingh


    Hi Friends,

    I want clear description about heartbeat over IP aias, IPAT with & without alias.

    Please enlighten me on these topics.

    thanks
    #PowerHA-(Formerly-known-as-HACMP)-Technical-Forum
    #PowerHAforAIX


  • 2.  Re: HACMP Questions

    Posted Thu January 21, 2010 04:44 PM

    Originally posted by: Casey_B


    Hello Gupreet.

    That's a bit vague, and open ended.
    :)

    How's about I give you some starting information, and you come back with any specific questions that you have.

    1) IPAT means IP address takeover.
    It describes the feature of HA where a service label can be moved from one interface to another on the same node, or even to another interface on another node.

    2) There are two methodologies for implementing the IP address takeover. One is to replace the base ip address with the service label.
    This is called IPAT via replacement.

    In order to have a complete configuration, you will need to have one interface per service label to initially bring up the service label, and another interface
    available to receive the service label in case of fallover.

    There are also specific subnetting requirements. The boot ip label needs to be on the same subnet as the standby ip label.

    3) IPAT via aliasing moves the ip address by aliasing the service ip label on top of an already existing base address on an interface.
    A full configuration can be done with less interfaces, because multiple service labels can be placed on one interface.

    There are also specific subnetting requirements for this type of IPAT. Boot labels need to be on a seperate subnet from standby labels, which is different from service labels.

    4) Both of the subnetting requirements are due to the ability of the cluster node to detect failure of other interfaces in the cluster.
    Some of those restrictions can be removed if you allow HA to create a set of aliases specifically for heartbeating between the interfaces.

    Hope this gives you a start...

    Please respond with some more specific questions
    Casey
    #PowerHAforAIX
    #PowerHA-(Formerly-known-as-HACMP)-Technical-Forum