Topic
IC4NOTICE: developerWorks Community will be offline May 29-30, 2015 while we upgrade to the latest version of IBM Connections. For more information, read our upgrade FAQ.
6 replies Latest Post - ‏2014-08-25T13:14:29Z by LukeNumrych
Rizwan Ali Joo
Rizwan Ali Joo
26 Posts
ACCEPTED ANSWER

Pinned topic High Availability and Disaster Recovery

‏2014-08-20T20:22:43Z |

Hi, we have DB2 9.7.0.3 running on AIX 6, we have just configures HADR. During configuration we have not used the 

$ db2 update alternate server for database <db_name> using hostname <host_name/ip> port <port_number> 

command on any server, neither on Primary nor on Standby. Now, we want to test switchover and failover. Is it possible?

Thanks in Advance

  • Taml
    Taml
    5 Posts
    ACCEPTED ANSWER

    Re: High Availability and Disaster Recovery

    ‏2014-08-21T12:07:05Z  in response to Rizwan Ali Joo

    Rizwan,

    You only want to switch the role on the server or you expect the application to be routed to new primary ?

    Regards,

    Tamil

     

  • LukeNumrych
    LukeNumrych
    88 Posts
    ACCEPTED ANSWER

    Re: High Availability and Disaster Recovery

    ‏2014-08-21T14:17:11Z  in response to Rizwan Ali Joo

    The alternate server setting is only needed if you are going to use the Automated Client Reroute feature.  What it does is that it lets the client know that there is an alternate server that the client can use in case the primary server is down or unreachable.

    If you are not planning on using ACR, you do not need to configure this.

  • Rizwan Ali Joo
    Rizwan Ali Joo
    26 Posts
    ACCEPTED ANSWER

    Re: High Availability and Disaster Recovery

    ‏2014-08-22T09:34:16Z  in response to Rizwan Ali Joo

    Thank You #Taml and LuckNumrych.

    LuckNumrych, I got your point. but i want to understand what is the purpose of ACR?

    Thanks in Advance

    • LukeNumrych
      LukeNumrych
      88 Posts
      ACCEPTED ANSWER

      Re: High Availability and Disaster Recovery

      ‏2014-08-22T12:01:47Z  in response to Rizwan Ali Joo

      Like I wrote - it lets the client know that there is an alternate server (it could be the standby-turned-primary of a HADR cluster, but not necessarily) to which the client can connect in case the original primary server is not available.

      In the simplest scenario, it works like this:

      1. client connects to a primary server that has an alternate server defined
      2. client works with the primary server
      3. primary server becomes unavailable
      4. client tries working with the primary server but it is unable to do so
      5. client is AUTOMATICALLY REROUTED to the alternate server.

      There are several caveats to this, so make sure you read the documentation.