Problem failover a clustered application when when cluster failover

Discussion Problem failover a clustered application when when cluster failover

  • Tuesday, June 12, 2012 1:31 AM
     
     

    I conducted a failover test to test the failover of the active node to standby node by powering down the active node server. 

    The cluster service did successfully failed over to the standy node. Everything (such as network service, drive, etc.) was failed over successfully except the DB2 application in cluster. 

    Somehow, the DB2 service was marked as “Failed” in the Cluster Manager. 

    After some time (About 10min), the application miraculously started again.  We don’t know for sure if it retried or waited.

    Will there be a dependency on the clustered drive (M: and N:0) since DB2 database and log are located on them?

    I'd appreciate if someone can shed light on this issue, telling me the root cause.

    Much appreciated.

All Replies

  • Tuesday, June 12, 2012 9:12 AM
     
     

    Can you please collect cluster logs and share any obvious errors.

    Thanks

    Sudhir

  • Wednesday, June 13, 2012 3:29 AM
     
     

    Hi,

    As per Sudhir, we would really need to see the cluster log.

    Can you re-produce the issue and post the log for the time period that the issue was re-produced in?

    Thanks,

    Martin


    If you find my information useful, please rate it. :-)

  • Friday, June 15, 2012 6:04 AM
     
     

    Dear All,

    thanks for the reply. I will get the log to see if there is any abnoamality.