Deploying Multi-Pod Clusters

Prerequisites

  • To deploy TIBCO Product and Service Catalog in high-availability mode, read the hardware requirements at Hardware Requirements.
  • Choose a suitable database for this deployment. TIBCO Product and Service Catalog currently supports Oracle and PostgreSQL databases running on-premises or on the cloud.

Multi-pod cluster deployment on Kubernetes consists of the following broad steps:

(i) Create a namespace,

(ii) Create Kubernetes entities (such as storage class and persistent volume),

(iii) Deploy the database, TIBCO EMS, Apache Ignite, and

(iv) Deploy with TIBCO Product and Service Catalog or TIBCO Offer and Pricing Designer. For detailed steps, see the procedure.

Prerequisites

  • To deploy TIBCO Product and Service Catalog in High-availability mode, read the hardware requirements at Hardware Requirements
  • Choose a suitable database for this deployment. TIBCO Product and Service Catalog currently supports Oracle and PostgreSQL databases running on-premises or on the cloud.

Procedure

  1. Create a namespace for working with multi-pod TIBCO Product and Service Catalog deployment on Kubernetes.
    1. ensure that the following values are entered in fc-namespace.json:
      • Kind: Namespace
      • Name in metadata: tibco-fc-500
      • Name in labels: tibco-fc-500
    2. Run the command to create the namespace: 
    $ kubectl create -f fc-namespace.json
  2. Create the following entities in the specified order:
    No. Name Details
    i. Storage class Configuration details for fc-storageclass.json
    • Kind: StorageClass
    • Namespace: tibco-fc-500

    Command to create a storage class 

    $ kubectl create -f fc-storageclass.json
    ii. Persistent Volume Configuration details for fc-pv.json
    • Kind: PersistentVolume
    • Path in hostPath: This path must be already present in the worker node or it is created by Kubernetes after kubectl creates -f fc-pv.json. You must create the mounted directory mentioned in the fc-pv.json file and give all the permissions.
    • Name of the worker node: Name of the created worker node
    • Storage: Size of the Persistent Volume (10Gi for 10 gb)
    Note: In fc-pv.json, you must provide values for worker-node1 hostname and Kubernetes master hostname.
    Command to create a Persistent Volume 
    $ kubectl create -f fc-pv.json
    iii. Persistent Volume Claim Configuration details for fc-pvc.json
    • Kind: PersistentVolumeClaim
    • StorageClassName: Name of the storage class created
    • Storage: Size of the Persistent Volume Claim (upto 1 GB or more)
    Command to create a Persistent Volume Claim 
    $ kubectl create -f fc-pvc.json
  3. Deploy a supported database. You need the database details mentioned later in the procedure.
    Purpose Reference
    To deploy containerized PostgreSQL Database Deploying Containerized PostgreSQL Database
    To deploy on-premise PostgreSQL Database Deploying On-premise PostgreSQL Database
    To deploy containerized Oracle Database Deploying Containerized Oracle Database
    To deploy on-premise Oracle Database Deploying On-premise Oracle Database
  4. Deploy TIBCO Enterprise Message Service (TIBCO EMS)TIBCO Product and Service Catalog integrates with on-premises TIBCO EMS as well as cloud TIBCO EMS.
    Purpose Reference
    To deploy containerized Enterprise Message Service Deploying Containerized Enterprise Message Service
    To deploy on-premises Enterprise Message Service Deploying On-premises Enterprise Message Service
  5. Create a configuration map.
    1. Ensure that the following values are entered in fc-properties.properties:
      Property Value
      MQ_MDM_DB_TYPE POSTGRES/ ORACLE
      MQ_MDM_DB_HOST Database host / fc db Kubernetes service
      MQ_MDM_DB_PORT Database port
      MQ_MDM_DB_NAME Database service name as configured in tnsnames
      MQ_MDM_DB_SCHEMA Database schema name
      MQ_MDM_DB_MIN_CONN_COUNT 20
      MQ_MDM_DB_MAX_CONN_COUNT 250
      MQ_MDM_DB_IGN_MIN_CONN_COUNT 1
      MQ_MDM_DB_IGN_MAX_CONN_COUNT 10
      EMS_SERVER_URL tcp://host:port, if Enterprise Message Service is running on an on-premises setup. In this case, the host value is the hostname or IP address on which the Enterprise Message Service is running. If the Enterprise Message Service is containerized, then the host value is Enterprise Message Service service name or IP address.
      FC_SERVICE_HOST IP address of service where TIBCO Product and Service Catalog is deployed
      FC_SERVICE_PORT Port on which TIBCO Product and Service Catalog is running
      OPD_APP_DATA_VOLUME heavy/light
    2. Run the following command to create a config map: 
      $kubectl create cm fc-config -n tibco-fc-500 --from-env-file=<path to fc-properties.properties>
  6. To create secrets for a database and Enterprise Message Service, ensure that the following values are entered in fc-secret.json:
    • MQ_MDM_DB_USER=<base64 encoded database username>
    • MQ_MDM_DB_PASSWORD=<base64 encoded database password>
    • MQ_MDM_DB_ADMIN_USER=<base64 encoded database admin username>
    • MQ_MDM_DB_ADMIN_USER_PWD=<base64 encoded database admin password>
    • EMS_USER_NAME=<base 64 encoded ems user name>
    • EMS_PASSWORD=<base 64 encoded ems password>
  7. Deploy Apache Ignite.
    1. Ensure that the following values are entered in ignite.json:
      • Kind: Deployment
      • App in matchLabels: fc
      • Tier in matchLabels: ignite
      • Image: Docker registry URL of image with latest tag
      • Name in imagePullSecrets: Name of the registry credentials created to pull the images
    2. Run the following command to create fcdb: 
    $ kubectl create -f ignite.json
    Note: To change the JVM parameters of the Ignite server, update the entry-point.sh file.
  8. When deploying with TIBCO Product and Service Catalog, perform the following steps:
    1. In the tibco-fc-500 namespace, give the permissions in viewClusterRole to the service account located in the tibco-fc-500 namespace and named default. You must create rolebinding to access the TIBCO Product and Service Catalog instance in the cluster to create a Wildfly cluster. 
      $kubectl create rolebinding default-viewer --clusterrole=view
--serviceaccount=tibco-fc-500:default --namespace=tibco-fc-500
    2. Ensure that the following values are entered in fc.json:
      • Kind: Deployment
      • App in matchLabels: fc (this value must be the same as the value of the app in selector in fc-service.json)
      • Tier in matchLabels: instance (this value must be the same as the value of the tier in selector in fc-service.json)
      • Image: Docker registry url of image with the latest tag
      • Name in imagePullSecrets: Name of the registry credentials created to pull images
    3. Run the following command to create TIBCO Product and Service Catalog deployment: 
      $kubectl create -f fc.json
      Note: To change the JVM parameters of the Ignite server, other than the default ones, update the entry-point.sh file.
  9. Run the following command to create TIBCO Product and Service Catalog service: 
    $kubectl create -f fc-service.json
  10. Run the following command to create the configurator service kubectl create -f configurator-service.json: 
    $kubectl create -f configurator-service.json
  11. When deploying with Offer and Pricing Designer (OPD), perform the following steps:
    1. Ensure that the following values are entered in the opd.json:
      • Kind: Deployment
      • App in matchLabels: fc (this value must be the same as the value of the app in selector in opd-service.json)
      • Tier in matchLabels: opd (this value must be the same as the value of the tier in selector in opd-service.json)
      • Image: Docker registry url of image with the latest tag
      • Name in imagePullSecrets: name of the registry credentials created to pull images
      Run the following command to create TIBCO Offer and Pricing Designer: 
      $kubectl create -f opd.json
      Note: The existing topology of TIBCO Product and Service Catalog deployment is that it supports only a single pod of OPD. If OPD pod is scaled up, OPD features do not work as expected.
    2. Ensure that the following values are entered in the opd-service.json:
      • Kind: Service
      • Type: LoadBalancer
      • Ports: 
        [
         {
            "name": "http-opd",
            "protocol": "TCP",
            "port": 8070,
            "targetPort": 8070
         }
      ]
      Run the following command to create Offer and Pricing Designer service: 
      $kubectl create -f opd-service.json