Preparing storage for Content Cortex file storage areas and index areas

To store your Content Cortex-managed documents in your file system, provide a persistent volume claim (PVC) for the Content Platform Engine container to use during deployment. To use the full text indexing feature of the Content Search Services, you set up and configure an extra PVC to prepare for the Content Search Services container deployment.

To prepare storage for storage areas and index areas:

1. To prepare your file system storage to be used as an advanced storage area device, review the information in the following topics:
   • Replication models for advanced storage areas
   • Preparing advanced storage areas
2. To prepare your file system storage to be used as an advanced storage area device, use the following procedures: Preparing file servers for file storage areas
   Note:

   • Follow the suggested action of creating a subdirectory under a top-level directory for each advanced storage area device or file storage area you intend to create. If you do not follow this suggested action, and you later must create another top-level directory for an extra device or area, you must rearrange the storage or use other techniques to provide a single top-level directory. A single top-level directory is needed for the persistent volume definition that corresponds to the Content Platform Engine filestore volume for the Content Platform Engine container deployment.
   • The top-level directory and all of its subfolders must be owned by the user, recorded earlier as the value for cpe_os_user, whose numerical ID is 50001 and the group, recorded earlier as cpe_os_group, whose numerical ID is 50000. This permission setting allows the Content Platform Engine container to access the file system.
3. To prepare your file system storage to be used as an index area, use the following procedures: Preparing for IBM Content Search Services
   Note:

   • Create a subdirectory under a top-level directory for each object store you intend to enable for full text indexing. You might want to create a hierarchy where there is a top-level directory with subdirectories for each object store, then each object store subdirectory that has a subdirectory for each index area that will be created. If you do not follow this guidance, and you later must create another top-level directory for an index area, you need to rearrange the storage or use other techniques to provide a single top-level directory. A single top-level directory is needed for the persistent volume definition that corresponds to the index volume for content index data for the Content Search Services container deployment.
   • The top-level directory and all of its subfolders must be owned by the user, recorded earlier as the value for cpe_os_user, whose numerical ID is 50001 and the group, recorded earlier as cpe_os_group, whose numerical ID is 50000. This permission setting allows the Content Platform Engine container to access the file system.
4. Record for later use by the Content Cortex administrator any pre-created subdirectories that hold advanced storage area devices or index areas data.
   Regardless of the file system that is visible from the worker nodes, the content services components that operate as containers view the file system from an inside-the-container perspective. When using the Content Cortex administration tools and other configuration interfaces to provide information such as root paths or locations of files, the fully qualified path uses the mountPath value and not the actual path.

   Example 1: Advanced storage area device

   A marketing department requests a content services deployment. The top-level directory for the file storage areas to use with this deployment is created with a mount path of /data/FNdomain_Marketing.

   Two subdirectories are created as /data/FNdomain_Marketing/ASAdevice1 and /data/FNdomain_Marketing/ASAdevice2.

   The PV and PVC pair for the Content Platform Engine filestore volume that is used for the Content Platform Engine container deployment claims is /data/FNdomain_Marketing.

   The mountPath value for the Content Platform Engine file store volume is /opt/ibm/asa.

   When later creating an advanced storage area device object in the object store that needs a file storage area, the root path that is given by the Content Cortex administrator is /opt/ibm/asa/ASAdevice1.

   Example 2: Index area

   In the same content services deployment for the marketing department, the top-level directory for the index areas was created with a mount point of /indexes/FNdomain_Marketing.

   Two subdirectories are created as /indexes/FNdomain_Marketing/Docs1 and /indexes/FNdomain_Marketing/Docs2.

   The PV and PVC pair for the index volume for content index data that is used for the Content Search Services container deployment claims is /indexes/FNdomain_Marketing.

   The mountPath value for the index volume for content index data is /opt/ibm/indexareas.

   When later creating an index area in the object store that is enabled for full text indexing, the root path that is given by the Content Cortex administrator is /CSSIndex1_OS1/Docs1.

   More information on the inside mountPath values can be found in the topic on Creating volumes and folders for deployment.