Deploying IBM Cloud Pak for Knowledge on Pink Hat OpenShift Service on AWS

Amazon Internet Companies (AWS) clients who’re searching for a extra intuitive strategy to deploy and use IBM Cloud Pak for Knowledge (CP4D) on the AWS Cloud, can now use the Pink Hat OpenShift Service on AWS (ROSA).

ROSA is a totally managed service, collectively supported by AWS and Pink Hat. It’s managed by Pink Hat Web site Reliability Engineers and offers a pay-as-you-go pricing mannequin, in addition to a unified billing expertise on AWS.

With this, clients don’t handle the lifecycle of Pink Hat OpenShift Container Platform clusters. As an alternative, they’re free to give attention to growing new options and innovating sooner, utilizing IBM’s built-in knowledge and synthetic intelligence platform on AWS, to distinguish their enterprise and meet their ever-changing enterprise wants.

CP4D may also be deployed from the AWS Market with self-managed OpenShift clusters. That is ultimate for patrons with necessities, like Pink Hat OpenShift Knowledge Basis software program outlined storage, or preferring to handle their OpenShift clusters.

On this submit, we talk about methods to deploy CP4D on ROSA utilizing IBM-provided Terraform automation.

Cloud Pak for knowledge structure

Right here, we set up CP4D in a extremely obtainable ROSA cluster throughout three availability zones (AZs); with three grasp nodes, three infrastructure nodes, and three employee nodes.

Evaluation the AWS Areas and Availability Zones documentation and the areas the place ROSA is obtainable to decide on the perfect area on your deployment.

It is a public ROSA cluster, accessible from the web by way of port 443. When deploying CP4D in your AWS account, think about using a personal cluster (Determine 1).

Determine 1. IBM Cloud Pak for Knowledge on ROSA

We’re utilizing Amazon Elastic Block Retailer (Amazon EBS) and Amazon Elastic File System (Amazon EFS) for the cluster’s persistent storage. Evaluation the IBM documentation for details about supported storage choices.

Evaluation the AWS conditions for ROSA, and comply with the Safety finest practices in IAM documentation to guard your AWS account earlier than deploying CP4D.

Price

The prices related to utilizing AWS companies when deploying CP4D in your AWS account will be estimated on the pricing pages for the companies used.

Conditions

This weblog assumes familiarity with: CP4D, Terraform, Amazon Elastic Compute Cloud (Amazon EC2), Amazon EBS, Amazon EFS, Amazon Digital Non-public Cloud, and AWS Id and Entry Administration (IAM).

You’ll need the next earlier than getting began:

• Entry to an AWS account, with permissions to create the sources described within the set up steps part.
• An AWS IAM person, with the permissions described within the AWS conditions for ROSA documentation.
• Adequate AWS service quotas to deploy ROSA. You possibly can request service-quota will increase from the AWS console.
• An IBM entitlement API key: both a 60-day trial or an current entitlement.
• A bastion host to run the CP4D installer, with the next packages:

Set up steps

Full the next steps to deploy CP4D on ROSA:

1. First, allow ROSA on the AWS account. From the AWS ROSA console, click on on Allow ROSA, as in Determine 2.
   

   Determine 2. Enabling ROSA in your AWS account

2. Click on on Get began. Redirect to the Pink Hat web site, the place you’ll be able to register and acquire a Pink Hat ROSA token.
3. Navigate to the AWS IAM console. Create an IAM coverage named cp4d-installer-policy and add the next permissions:

   {
       "Model": "2012-10-17",
       "Assertion": [
           {
               "Effect": "Allow",
               "Action": [
                   "autoscaling:*",
                   "cloudformation:*",
                   "cloudwatch:*",
                   "ec2:*",
                   "elasticfilesystem:*",
                   "elasticloadbalancing:*",
                   "events:*",
                   "iam:*",
                   "kms:*",
                   "logs:*",
                   "route53:*",
                   "s3:*",
                   "servicequotas:GetRequestedServiceQuotaChange",
                   "servicequotas:GetServiceQuota",
                   "servicequotas:ListServices",
                   "servicequotas:ListServiceQuotas",
                   "servicequotas:RequestServiceQuotaIncrease",
                   "sts:*",
                   "support:*",
                   "tag:*"
               ],
               "Useful resource": "*"
           }
       ]
   }

4. Subsequent, let’s create an IAM person from the AWS IAM console, which can be used for the CP4D set up:
   a. Specify a reputation, like ibm-cp4d-bastion.
   b. Set the credential sort to Entry key – Programmatic entry.
   c. Connect the IAM coverage created in Step 3.
   d. Obtain the .csv credentials file.
5. From the Amazon EC2 console, create a brand new EC2 key pair and obtain the personal key.
6. Launch an Amazon EC2 occasion from which the CP4D installer is launched:
   a. Specify a reputation, like ibm-cp4d-bastion.
   b. Choose an occasion sort, similar to t3.medium.
   c. Choose the EC2 key pair created in Step 4.
   d. Choose the Pink Hat Enterprise Linux 8 (HVM), SSD Quantity Kind for 64-bit (x86) Amazon Machine Picture.
   e. Create a safety group with an inbound rule that enables connection. Limit entry to your personal IP deal with or an IP vary out of your group.
   f. Go away all different values as default.
7. Hook up with the EC2 occasion by way of SSH utilizing its public IP deal with. The remaining set up steps can be initiated from it.
8. Set up the required packages:

   $ sudo yum replace -y
   $ sudo yum set up git unzip vim wget httpd-tools python38 -y
   
   $ sudo ln -s /usr/bin/python3 /usr/bin/python
   $ sudo ln -s /usr/bin/pip3 /usr/bin/pip
   $ sudo pip set up pyyaml
   
   $ curl "https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip" -o "awscliv2.zip"
   $ unzip awscliv2.zip
   $ sudo ./aws/set up
   
   $ wget "https://github.com/stedolan/jq/releases/obtain/jq-1.6/jq-linux64"
   $ chmod +x jq-linux64
   $ sudo mv jq-linux64 /usr/native/bin/jq
   
   $ wget "https://mirror.openshift.com/pub/openshift-v4/purchasers/ocp/4.10.15/openshift-client-linux-4.10.15.tar.gz"
   $ tar -xvf openshift-client-linux-4.10.15.tar.gz
   $ chmod u+x oc kubectl
   $ sudo mv oc /usr/native/bin
   $ sudo mv kubectl /usr/native/bin
   
   $ sudo yum set up -y yum-utils
   $ sudo yum-config-manager --add-repo $ https://rpm.releases.hashicorp.com/RHEL/hashicorp.repo
   $ sudo yum -y set up terraform
   
   $ sudo subscription-manager repos --enable=rhel-7-server-extras-rpms
   $ sudo yum set up -y podman

9. Configure the AWS CLI with the IAM person credentials from Step 4 and the specified AWS area to put in CP4D:

   $ aws configure
   
   AWS Entry Key ID [None]: AK****************7Q
   AWS Secret Entry Key [None]: vb************************************Fb
   Default area identify [None]: eu-west-1
   Default output format [None]: json

10. Clone the next IBM GitHub repository:
    https://github.com/IBM/cp4d-deployment.git

    $ cd ~/cp4d-deployment/managed-openshift/aws/terraform/

11. For the aim of this submit, we enabled Watson Machine Studying, Watson Studio, and Db2 OLTP companies on CP4D. Use the instance on this step to create a Terraform variables file for CP4D set up. Allow CP4D companies required on your use case:

    area			= "eu-west-1"
    tenancy			= "default"
    access_key_id 		= "your_AWS_Access_key_id"
    secret_access_key 	= "your_AWS_Secret_access_key"
    
    new_or_existing_vpc_subnet	= "new"
    az				= "multi_zone"
    availability_zone1		= "eu-west-1a"
    availability_zone2 		= "eu-west-1b"
    availability_zone3 		= "eu-west-1c"
    
    vpc_cidr 		= "10.0.0.0/16"
    public_subnet_cidr1 	= "10.0.0.0/20"
    public_subnet_cidr2 	= "10.0.16.0/20"
    public_subnet_cidr3 	= "10.0.32.0/20"
    private_subnet_cidr1 	= "10.0.128.0/20"
    private_subnet_cidr2 	= "10.0.144.0/20"
    private_subnet_cidr3 	= "10.0.160.0/20"
    
    openshift_version 		= "4.10.15"
    cluster_name 			= "your_ROSA_cluster_name"
    rosa_token 			= "your_ROSA_token"
    worker_machine_type 		= "m5.4xlarge"
    worker_machine_count 		= 3
    private_cluster 			= false
    cluster_network_cidr 		= "10.128.0.0/14"
    cluster_network_host_prefix 	= 23
    service_network_cidr 		= "172.30.0.0/16"
    storage_option 			= "efs-ebs" 
    ocs 				= { "allow" : "false", "ocs_instance_type" : "m5.4xlarge" } 
    efs 				= { "allow" : "true" }
    
    accept_cpd_license 		= "settle for"
    cpd_external_registry 		= "cp.icr.io"
    cpd_external_username 	= "cp"
    cpd_api_key 			= "your_IBM_API_Key"
    cpd_version 			= "4.5.0"
    cpd_namespace 		= "zen"
    cpd_platform 			= "sure"
    
    watson_knowledge_catalog 	= "no"
    data_virtualization 		= "no"
    analytics_engine 		= "no"
    watson_studio 			= "sure"
    watson_machine_learning 	= "sure"
    watson_ai_openscale 		= "no"
    spss_modeler 			= "no"
    cognos_dashboard_embedded 	= "no"
    datastage 			= "no"
    db2_warehouse 		= "no"
    db2_oltp 			= "sure"
    cognos_analytics 		= "no"
    master_data_management 	= "no"
    decision_optimization 		= "no"
    bigsql 				= "no"
    planning_analytics 		= "no"
    db2_aaservice 			= "no"
    watson_assistant 		= "no"
    watson_discovery 		= "no"
    openpages 			= "no"
    data_management_console 	= "no"

12. Save your file, and launch the instructions beneath to put in CP4D and observe progress:

    $ terraform init -input=false
    $ terraform apply --var-file=cp4d-rosa-3az-new-vpc.tfvars 
       -input=false | tee terraform.log

13. The set up runs for 4 or extra hours. As soon as set up is full, the output contains (as in Determine 3):
    a. Instructions to get the CP4D URL and the admin person password
    b. CP4D admin person
    c. Login command for the ROSA cluster

Determine 3. CP4D set up output

Validation steps

Let’s confirm the set up!

1. Log in to your ROSA cluster utilizing your cluster-admin credentials.

   $ oc login https://api.cp4dblog.17e7.p1.openshiftapps.com:6443 --username cluster-admin --password *****-*****-*****-*****

2. Provoke the next command to get the cluster’s console URL (Determine 4):

   $ oc whoami --show-console

   

   Determine 4. ROSA console URL

3. Run the instructions on this step to retrieve the CP4D URL and admin person password (Determine 5).

   $ oc extract secret/admin-user-details 
     --keys=initial_admin_password --to=- -n zen
   $ oc get routes -n zen

   

   Determine 5. Retrieve the CP4D admin person password and URL

4. Provoke the next instructions to have the CP4D workloads in your ROSA cluster (Determine 6):

   $ oc get pods -n zen
   $ oc get deployments -n zen
   $ oc get svc -n zen 
   $ oc get pods -n ibm-common-services 
   $ oc get deployments -n ibm-common-services
   $ oc get svc -n ibm-common-services
   $ oc get subs -n ibm-common-services

   

   Determine 6. Checking the CP4D pods operating on ROSA

5. Log in to your CP4D internet console utilizing its URL and your admin password.
6. Increase the navigation menu. Navigate to Companies > Companies catalog for the obtainable companies (Determine 7).
   

   Determine 7. Navigating to the CP4D companies catalog

7. Discover that the companies set as “enabled” correspond along with your Terraform definitions (Determine 8).
   

   Determine 8. Companies enabled in your CP4D catalog

Congratulations! You’ve efficiently deployed IBM CP4D on Pink Hat OpenShift on AWS.

Put up set up

Check with the IBM documentation on establishing companies, if you might want to allow further companies on CP4D.

When putting in CP4D on productive environments, please assessment the IBM documentation on securing your atmosphere. Additionally, the Pink Hat documentation on establishing id suppliers for ROSA is informative. You can even think about enabling auto scaling on your cluster.

Cleanup

Hook up with your bastion host, and run the next steps to delete the CP4D set up, together with ROSA. This step avoids incurring future costs in your AWS account.

$ cd ~/cp4d-deployment/managed-openshift/aws/terraform/
$ terraform destroy -var-file="cp4d-rosa-3az-new-vpc.tfvars"

Should you’ve skilled any failures in the course of the CP4D set up, run these subsequent steps:

$ cd ~/cp4d-deployment/managed-openshift/aws/terraform
$ sudo cp installer-files/rosa /usr/native/bin
$ sudo chmod 755 /usr/native/bin/rosa
$ Cluster_Name=`rosa checklist clusters -o yaml | grep -w "identify:" | lower -d ':' -f2 | xargs`
$ rosa take away cluster --cluster=${Cluster_Name}
$ rosa logs uninstall -c ${Cluster_Name } –watch
$ rosa init --delete-stack
$ terraform destroy -var-file="cp4d-rosa-3az-new-vpc.tfvars"

Conclusion

In abstract, we explored how clients can reap the benefits of a totally managed OpenShift service on AWS to run IBM CP4D. With this implementation, clients can give attention to what’s vital to them, their workloads, and their clients, and fewer on managing the day-to-day operations of managing OpenShift to run CP4D.

Take a look at the IBM Cloud Pak for Knowledge Simplifies and Automates How You Flip Knowledge into Insights weblog to discover ways to use CP4D on AWS to unlock the worth of your knowledge.

Extra sources