Oracle® Solaris Cluster Data Service for Oracle Business Intelligence Enterprise Edition Guide

Exit Print View

Updated: July 2014, E39576–01
 
 

BI Cluster Controller Configuration

This section describes how to configure the BI Cluster Controller and create the appropriate Oracle Solaris Cluster resource groups and resources to support highly available solutions.

How to Define a Highly Available BI Cluster Controller Configuration


Note -  The failover BI Cluster Controller resource groups provide the “glue” between the multi-master BI Presentation Services and the multi-master BI Servers and multi-master Schedulers.

Before You Begin

Before you create the BI Cluster Controller resource for a failover configuration, you might need to edit the following file:

  • /etc/netmasks – Ensure that the /etc/netmasks file has IP-address subnet and netmask entries for all logical hostnames. If necessary, edit the /etc/netmasks file to add any missing entries.

  • BI_INSTALL_DIRECTORY/server/Config/NQClusterConfig.INI – Verify the following settings:

    • ENABLE_CONTROLLER must be set to YES.

    • Primary and secondary cluster controllers are defined to listen on logical hosts, for example:

      PRIMARY_CONTROLLER = "red-lh2";
      SECONDARY_CONTROLLER = "red-lh3"; 
    • Servers are set to a list of the nodes that are defined in the multi-master resource group, for example:

      SERVERS = "phys-red1","phys-red2"; 
    • One of the nodes in the multi-master resource group is set to be the master server, for example:

      MASTER_SERVER = "phys-red1";
  1. On each node, verify that the BI Cluster Controller can start.
    1. Export the necessary parameters.

      The following example is a single command. The line breaks are added for readability.

      SC_LHOSTNAME=fo-bi-pcc-lh;LD_PRELOAD_32=${LD_PRELOAD_32}
      :/usr/cluser/lib/libschost.so.1;LD_PRELOAD_64=${LD_PRELOAD_64}
      :/usr/cluster/lib/64/libschost.so.1;
      export SC_LHOSTNAME LD_PRELOAD_32 LD_PRELOAD_64
    2. Start the BI Cluster Controller.
      # run-ccs.sh start
    3. Confirm that the BI Cluster Controller process is running.
      # ps -ef | grp nqscheduler
  2. Create a resource group for the primary BI Cluster Controller.
    # clresourcegroup create -n phys-red1 fo-bi-pcc-rg 

    Note -  Although the failover resource group can contain more than one node, the list of nodes for the primary BI Cluster Controller resource group and the list of nodes for the secondary BI Cluster Controller resource group cannot intersect.
  3. Create the logical hostname resource for the primary BI Cluster Controller.
    # clreslogicalhostname create -h red-lh2 -g fo-bi-pcc-rg fo-bi-pcc-lh-rs 
  4. Create the resource for the primary BI Cluster Controller.
    # clresource create -t ORCL.obiee_cluster_controller \
    -g fo-bi-pcc-rg \
    -p BI_INSTALL_DIRECTORY=/export/home/obiee/OracleBI \
    -p BI_DATA_DIRECTORY=/export/home/obiee/OracleBIData \
    -p Run_64_bit=TRUE \
    -p CLUSTER_CONTROLLER_ROLE=PRIMARY \
    -p Resource_dependencies=fo-bi-pcc-lh-rs \
    fo-bi-pcc-rs 
  5. Create a resource group for the secondary BI Cluster Controller.
    # clresourcegroup create -n phys-red2 fo-bi-scc-rg

    Note -  Although the failover resource group can contain more than one node, the list of nodes for the primary BI Cluster Controller resource group and the list of nodes for the secondary BI Cluster Controller resource group cannot intersect.
  6. Create the logical hostname resource for the secondary BI Cluster Controller.
    # clreslogicalhostname create -h red-lh3 -g fo-bi-scc-rg fo-bi-scc-lh-rs 
  7. Create the resource for the secondary BI Cluster Controller.
    # clresource create -t ORCL.obiee_cluster_controller \
    -g fo-bi-scc-rg \
    -p BI_INSTALL_DIRECTORY=/export/home/obiee/OracleBI \
    -p BI_DATA_DIRECTORY=/export/home/obiee/OracleBIData \
    -p Run_64_bit=TRUE \
    -p CLUSTER_CONTROLLER_ROLE=SECONDARY \
    -p Resource_dependencies=fo-bi-scc-lh-rs \
    fo-bi-scc-rs 
  8. Bring the BI Cluster Controller resource groups online.
    # clresourcegroup online -eM fo-bi-pcc-rg fo-bi-scc-rg 
  9. Verify the resource status.
    # clresource status -g fo-bi-pcc-rg,fo-bi-scc-rg
    === Cluster Resources ===
    Resource Name        Node Name	State 	Status Message
    -------------        ---------    ----- 	------	--------
    fo-bi-pcc-lh-rs          phys-red1 	Online 	Online - LogicalHostname online.
    fo-bi-pcc-rs            phys-red1 	Online 	Online - Service is online.
    fo-bi-scc-lh-rs          phys-red2 	Online 	Online - LogicalHostname online.
    fo-bi-scc-rs            phys-red2 	Online 	Online - Service is online.