Configuring Content Search Services

After a move to the Cloud Pak or container environment, you update your index areas to account for the environment changes.

Procedure

To configure Content Search Services after moving to the Cloud Pak:

  1. For each existing index area, edit the index areas root path values to reflect the mountPath for the Content Search Services containers:
    1. Log in to the Administration Console for Content Platform Engine using the credentials of the object store owner. If you use different credentials, you might not have the required permissions for the object store.
    2. In the Administration Console for Content Platform Engine, navigate to Domain > Object stores > object_store > Administrative > Index Areas.
    3. Choose an index area to edit.
    4. On the General tab, edit the property Root Directory based on the Content Search Service index area volume mapping to the mountPath/opt/ibm/indexareas.
      For example, if an index area IndexArea1 exists at /data/Marketing_Indexes/IndexArea1 and the Content Search Services index store volume claim is to /data/Marketing_Indexes, the new value for the Root Directory is /opt/ibm/indexareas/IndexArea1.
    5. Repeat these steps for every index area in the FileNet P8 domain.
  2. Change the ownership of the existing index areas so that they can be accessed by the Content Search Services services that are deployed as containers.
    Modify the index areas parent directory and all children to have ownership of 65534. Use a command like the following example, where the sample parent directory is /data/Marketing_Indexes:
    chgrp -R 65534 /data/Marketing_Indexes

    If a single parent directory for all index areas does not exist, create the parent directory in the file system and reference from it to the pre-existing data. As an alternative, you can also move the index areas so that they are all located under a single parent directory.

  3. If your container deployment exists alongside an on-premises deployment, modify the file system that holds the index areas to provide a reference for the Content Search Service services that are deployed as Java™ applications.
    The reference enables the services to see the index area parent directory as /opt/ibm/indexareas.
  4. Modify the existing FileNet P8 domain configuration to incorporate the newly deployed Content Search Service services:
    1. Obtain the configuration details for the new Content Search Services services in the container deployment:
      1. From the command line, use your Content Search Services pod name and run the following command:
        kubectl exec -it pod pod_name bash
      2. From the command line, change to the /opt/IBM/ContentSearchServices/CSS_Server/bin directory, and run the following command to get the authentication token:
        ./configTool.sh printToken -configPath ../config

        Make a note of the token so you can use it during the configuration.

      3. Exit out of the pod.
      4. From the command line, run the following command to get information about the pod:
        kubectl describe pod pod_name

        Make a note of the IP and port (8199) so you can use it during the configuration.

    2. Configure a text search server for each IBM Content Search Services deployment by navigating to Global Configuration > Administration > Text Search Servers > New Text Search Server, and complete the wizard:

      Server state: Ensure all Content Search Services servers in the containerized environment are configured for Dual: IndexAndSearch mode.

      Server parameters: Set the Host name and Port values to the pod IP and the css ssl port as seen from the container. Add the authentication token. This encryption key is used to encrypt the password during text index backup and restore operations.

      Communication security: Check Enable use of the Secure Socket Layer (SSL) protocol. Uncheck Validate the SSL server certificate and Validate the SSL certificate host.

      Server association: Set the Affinity Group value to the same affinity group that is set for your existing on-premises CSS server.

    3. Save the server settings.
    4. Repeat the preceding steps for all the newly deployed Content Search Service services.
  5. Expand Index Areas, and update the path to point to the container's index area. To find the path, run the following command:
    kubectl exec -it pod_name bash

    Examine the output and look for the path value for index areas, for example, opt/ibm/indexareas.

  6. Alter any affinity groups that might exist to incorporate the additional text search servers. Navigate to Domain > Global Configuration > Administration > Text Search Servers, and choose a text search server to edit.
    1. Go to Domain > Global Configuration > Administration > Text Search Servers, and choose a text search server to edit.
    2. Ensure that one or more text search servers created for the containerized Content Search Services deployments are assigned as members in existing affinity groups.
    3. Remove the old text search servers from any affinity groups they might belong to.
    4. Leave the old text search servers disabled and schedule them for future deletion.
  7. Re-enable text indexing for the object stores.