Control-M/Enterprise Manager Configuration Agent cluster resource (process emmaintag.exe) is not stopped on the active node of a Windows Cluster during failover to another node

Version 1
    Share:|

    This document contains official content from the BMC Software Knowledge Base. It is automatically updated when the knowledge article is modified.


    PRODUCT:

    Control-M/Enterprise Manager


    COMPONENT:

    Control-M/Enterprise Manager


    APPLIES TO:

    Control-M/Enterprise Manager installed on Windows Cluster with dedicated PostgreSQL database instance



    PROBLEM:

    The Control-M/Enterprise Manager is installed on a windows cluster with a dedicated PostgreSQL database instance and all cluster resources are active on a node of the cluster.
    When you want to move the application to another node of the cluster, you bring all resources offline from the active node and bring them online on the destination node.
    On the original active node the Configuration Agent resource remains active and the related process emmaintag.exe remains running.

    In the maint_diag log file the following messages can appear:

    emdb    EMDB::DLDBErrorHandlerImp(thread 11020)    errorCode = -6
    emdb    EMDB::DLDBErrorHandlerImp(thread 11020)    message   = DL_STAT_ERR_CONNECTION_CREATEsql, code:0, state: 57P03-org.postgresql.util.PSQLException: FATAL: the database system is shutting down
    emdb    EMDB::DLDBErrorHandlerImp(thread 11020)    vendorMessage1 = org.postgresql.util.PSQLException: FATAL: the database system is shutting down
    emdb    EMDB::DLDBErrorHandlerImp(thread 11020)    isSql   = 1

     


    CAUSE:

    The cause of this problem is that the database instance is shut down before the Configuration Agent. So the Configuration Agent is not able to update the status of components in the DB and it hangs.


    SOLUTION:

    A CAR has been created to address this issue :
    CAR00185505 - Control-M/EM on Windows cluster machines with dedicated Postgres, fails to stop the Config Agent service resource, when failing over to the target node, resulting with the failure of the failover activity.

    Until it is implemented, please perform the following steps:

       
    1. add a dependency to the PostgreSQL database cluster resource in the Configuration Agent cluster resource
    2.  
    3. perform the failover to the destination node
    4.  
    5. remove the dependency created at point 1.
    The removal of the dependency is necessary because under normal operational circumstances, Configuration Agent must remain running even when the DB is down to send monitoring data to Configuration Manager. 
      

     


    Article Number:

    000157291


    Article Type:

    Solutions to a Product Problem



      Looking for additional information?    Search BMC Support  or  Browse Knowledge Articles