Starting the target environment

Start the new environment, including the deployment manager, managed node or nodes, and servers.
Figure 1. Sample environment after the target is started. The source environment is not running. The target can read from the databases.
The details of the diagram are provided in the figure caption.

Before you begin

Before you start the target environment, make sure that you followed all the instructions for upgrading your databases.

Procedure

  1. Manually drop the existing messaging engine tables before you start the target deployment environment.
    Tip: The messaging engine table names use the SIB prefix.
  2. If your database type is DB2 for z/OS, manually create new messaging engine tables after you drop the existing tables. For more information, see Creating messaging engine tables manually.
  3. Start the target deployment environment:
    Important: Start only one Cognos® cluster member server and wait for the Cognos service to be active before you start the other Cognos cluster member servers and IBM® Business Monitor clusters. Do not use the Ripplestart option on the cluster where the Cognos service is running.
    1. Start the deployment manager.
    2. Start the node agents.
    3. Start one cluster member and wait until the Cognos service has started. In a one cluster topology, any cluster member will do. In a three cluster topology, start one of the cluster members in the support cluster.
    4. Start the clusters.

    Because IBM Cognos Business Intelligence updates its content store database automatically and creates a runtime copy of the application binary files when it first starts, the first startup might take extra time.

  4. Check to make sure that all clusters, servers, and messaging engines are started.
    Note: You can ignore an error similar to the following error in the SystemOut.log; it will be gone after monitor model migration.
    [8/6/14 17:05:11:535 GMT+08:00] 00000c56 AlarmListener E   SCHD0063E: A task with ID 51 (BPELProcessMonitoringModel) failed to run on Scheduler DataServicesScheduler (sched/wbm/DataServicesScheduler) because of an exception: com.ibm.ws.scheduler.exception.TaskException: SCHD0137E: Unable to create EJB instance for TaskHandler: java.rmi.RemoteException: CORBA TRANSIENT 0x4942fe07 No; nested exception is:
org.omg.CORBA.TRANSIENT: initial and forwarded IOR inaccessible  vmcid: IBM  minor code: E07  completed: No SCHD0140I: EJB information:  Host: bxv9v316 (9.110.186.165), Port: 9,900, J2EE component: IBM_WBM_DATA_SERVICES#DataServicesSchedulerEJB.jar#ModelRootTaskHandler.
  5. Check the IBM Cognos BI log file to make sure that IBM Cognos BI started correctly. The log file is in profile_root\cognos\server_name\logs\cogserver.log. Check for an entry that proves that IBM Cognos BI successfully migrated its content store, for example:
    9.37.25.66:9080 6968 2014-02-27 12:02:11.308 -5 Thread-77 CM 5000 1 Audit.RTUsage.cms.CM StartService ContentManagerService Success CM-SYS-5090 Content Manager build 10.2.5000.330 started (10.2.5000.330;20130502141002, schema version 7.0019, implementation: CMDbStore - Java CMCache).
    Note: If the IBM Cognos BI log does not exist, you might see an error like the following error: 
    SRVE0100E: Uncaught init() exception created by servlet cfgss in application IBM Cognos: java.lang.UnsatisfiedLinkError: JCAM_Crypto_JNI (libCCLCore.so: cannot open shared object file: No such file or directory)
    To resolve the problem, restart the IBM Cognos BI server.
Parent topic: Migrating to IBM Business Monitor V8.5.7 on Windows