Before installing IBM Cloud Pak for Integration (CP4I) on an OpenShift Cluster , an administrator needs to verify that the cluster meets the infrastructure prerequisites such as CPU, memory, and overall resource availability.

The most effective way to check real-time resource utilization and capacity is by using the oc adm top nodes command.

This command provides a summary of CPU and memory usage for all nodes in the cluster.

It is part of the OpenShift administrator CLI ( oc adm ) , which is designed for cluster-wide management.

The information is fetched from Metrics Server or Prometheus , giving accurate real-time data.

It helps administrators assess whether the cluster has sufficient resources to support CP4I deployment.

Why is oc adm top nodes the correct answer? Example usage: oc adm top nodes

Example Output: scss

Copy

NAME CPU (cores) CPU% MEMORY (bytes) MEMORY%

worker-node- 1 1500 m 30% 8 Gi 40%

worker-node- 2 1200 m 25% 6 Gi 30%

Why the Other Options Are Incorrect? Option

Explanation

Correct?

A. Use the oc cluster status command to dump the node capacity and utilization statistics for all the nodes.

❌ Incorrect – The oc cluster status command does not exist in OpenShift CLI. Instead, oc status gives a summary of the cluster but does not provide node resource utilization .

❌

B. Use the oc get nodes -o wide command to obtain the capacity and utilization statistics of memory and compute of the nodes.

❌ Incorrect – The oc get nodes -o wide command only shows basic node details like OS, internal/external IPs, and roles. It does not display CPU or memory utilization .

❌

C. Use the oc describe nodes command to dump the node capacity and utilization statistics for all the nodes.

❌ Incorrect – The oc describe nodes command provides detailed information about a node , but not real-time resource usage . It lists allocatable resources but does not give current utilization stats.

❌

Final Answer: ✅ D. Use the oc adm top nodes command to obtain the capacity and utilization statistics of memory and compute of the nodes.

IBM Cloud Pak for Integration - OpenShift Infrastructure Requirements

Red Hat OpenShift CLI Reference - oc adm top nodes

Red Hat OpenShift Documentation - Resource Monitoring

IBM Cloud Pak for Integration (CP4I) v2021.2 Administration References:

Correct Ordered Steps to Delete IBM MQ from IBM Cloud Pak for Integration (CP4I):

1️ ⃣ Log in to your OpenShift cluster ' s web console.

Access the OpenShift web console to manage resources and installed operators.

2️ ⃣ Select Operators from Installed Operators in a project containing Queue Managers.

Navigate to the Installed Operators section and locate the IBM MQ Operator in the project namespace where queue managers exist.

3️ ⃣ Delete Queue Managers.

Before uninstalling the operator, delete any existing IBM MQ Queue Managers to ensure a clean removal.

4️ ⃣ Uninstall the Operator.

Finally, uninstall the IBM MQ Operator from OpenShift to complete the deletion process.

To properly delete IBM MQ from IBM Cloud Pak for Integration (CP4I), the steps must be followed in the correct order:

Logging into OpenShift Web Console – This step provides access to the IBM MQ Operator and related resources.

Selecting the Installed Operator – Ensures the correct project namespace and MQ resources are identified.

Deleting Queue Managers – Queue Managers must be removed before uninstalling the operator; otherwise, orphaned resources may remain.

Uninstalling the Operator – Once all resources are removed, the MQ Operator can be uninstalled cleanly.

IBM Cloud Pak for Integration (CP4I) v2021.2 Administration References:

IBM MQ in Cloud Pak for Integration

Managing IBM MQ Operators in OpenShift

Uninstalling IBM MQ on OpenShift

In IBM App Connect , which is part of IBM Cloud Pak for Integration (CP4I) , callable flows enable integration between different environments, including on-premises systems and cloud-based integration solutions deployed in an OpenShift cluster.

To facilitate this connectivity, two critical resources are used:

The Connectivity Agent acts as a bridge between cloud-hosted App Connect instances and on-premises integration servers.

It enables secure bidirectional communication by allowing callable flows to connect between cloud-based and on-premise integration servers.

This is essential for hybrid cloud integrations , where some components remain on-premises for security or compliance reasons.

The Routing Agent directs incoming callable flow requests to the appropriate App Connect integration server based on configured routing rules .

It ensures low-latency and efficient message routing between cloud and on-premise systems, making it a key component for hybrid integrations .

1. Connectivity Agent ( ✅ Correct Answer) 2. Routing Agent ( ✅ Correct Answer)

Why the Other Options Are Incorrect? Option

Explanation

Correct?

A. Sync server

❌ Incorrect – There is no " Sync Server " component in IBM App Connect. Synchronization happens through callable flows, but not via a " Sync Server " .

❌

C. Kafka sync

❌ Incorrect – Kafka is used for event-driven messaging , but it is not required for callable flows between cloud and on-premises environments.

❌

D. Switch server

❌ Incorrect – No such component called " Switch Server " exists in App Connect.

❌

Final Answer: ✅ B. Connectivity agent ✅ E. Routing agent

IBM App Connect - Callable Flows Documentation

IBM Cloud Pak for Integration - Hybrid Connectivity with Connectivity Agents

IBM App Connect Enterprise - On-Premise and Cloud Integration

IBM Cloud Pak for Integration (CP4I) v2021.2 Administration References:

IBM Cloud Pak for Integration (CP4I) licensing follows Virtual Processor Core (VPC)-based pricing , where licensing requirements differ between production and non-production environments .

For non-production environments , IBM typically requires half the number of VPC licenses compared to production environments when deploying components like API Connect, App Connect Enterprise, and IBM MQ .

This 50% reduction applies because IBM offers a non-production environment discount , which allows enterprises to use fewer VPCs for testing, development, and staging while still maintaining functionality.

IBM provides reduced VPC license requirements for non-production environments to lower costs.

The licensing ratio is generally 1:2 (Non-Production:Production) , meaning the non-production environment will require half the licenses compared to production.

This policy is commonly applied to major CP4I components, including:

IBM API Connect

IBM App Connect Enterprise

IBM MQ

A. The same amount → Incorrect

Non-production environments typically require half the VPC licenses, not the same amount .

C. More than half as many → Incorrect

IBM’s standard licensing policy offers at least a 50% reduction , so this is not correct.

D. More information is needed to determine the cost. → Incorrect

While pricing details depend on contract terms, IBM has a standard non-production licensing policy , making it predictable.

Why Answer B is Correct? Explanation of Incorrect Answers:

IBM Cloud Pak for Integration Licensing Guide

IBM Cloud Pak VPC Licensing

IBM MQ Licensing Details

IBM API Connect Licensing

IBM App Connect Enterprise Licensing

IBM Cloud Pak for Integration (CP4I) v2021.2 Administration References:

When installing a new instance of IBM Cloud Pak for Integration (CP4I) Operations Dashboard , several prerequisites must be met. The correct answers are B and D based on IBM Cloud Pak for Integration v2021.2 requirements.

B. A pull secret from IBM Entitled Registry must exist in the namespace containing an entitlement key.

The IBM Entitled Registry hosts the necessary container images required for CP4I components, including Operations Dashboard.

Before installation, you must create a pull secret in the namespace where CP4I is installed. This secret must include your IBM entitlement key to authenticate and pull images.

Command to create the pull secret:

Correct Answers: oc create secret docker-registry ibm-entitlement-key \

--docker-server=cp.icr.io \

--docker-username= cp \

--docker-password= < your_entitlement_key > \

--namespace= < your_namespace >

IBM Reference: IBM Entitled Registry Setup

D. The vm.max_map_count sysctl setting on worker nodes must be higher than the operating system default.

The Operations Dashboard relies on Elasticsearch , which requires an increased vm.max_map_count setting for better performance and stability.

The default Linux setting ( 65530 ) is too low. It needs to be at least 262144 to avoid indexing failures.

To update this setting permanently, run the following command on worker nodes:

sudo sysctl -w vm.max_map_count=262144

IBM Reference: Elasticsearch System Requirements

A. For shared data, a storage class that provides ReadWriteOnce (RWO) access mode of at least 100 MB is required. (Incorrect)

While persistent storage is required, the Operations Dashboard primarily uses Elasticsearch , which typically requires ReadWriteOnce (RWO) or ReadWriteMany (RWX) block storage . However, the 100 MB storage requirement is incorrect, as Elasticsearch generally requires gigabytes of storage , not just 100 MB.

IBM Recommendation: Typically, Elasticsearch requires at least 10 GB of persistent storage for logs and indexing.

C. If the OpenShift Container Platform Ingress Controller pod runs on the host network, the default namespace must be labeled with network.openshift.io/controller-group: ingress to allow traffic to the Operations Dashboard. (Incorrect)

While OpenShift’s Ingress Controller must be configured correctly, this specific label requirement applies to some specific OpenShift configurations but is not a mandatory prerequisite for Operations Dashboard installation.

Instead, route-based access and appropriate network policies are required to allow ingress traffic.

E. For storing tracing data, a block storage class that provides ReadWriteMany (RWX) access mode and 10 IOPS of at least 10 GB is required. (Incorrect)

Tracing data storage does require persistent storage , but block storage does not support RWX mode in most environments.

Instead, file-based storage with RWX access mode (e.g., NFS) is typically used for OpenShift deployments needing shared storage.

Explanation of Incorrect Options:

IBM Cloud Pak for Integration Operations Dashboard Installation Guide

Setting Up IBM Entitled Registry Pull Secret

Elasticsearch System Configuration - vm.max_map_count

OpenShift Storage Documentation

IBM Cloud Pak for Integration (CP4I) v2021.2 Administration References:

Final Answer: ✅ B. A pull secret from IBM Entitled Registry must exist in the namespace containing an entitlement key. ✅ D. The vm.max_map_count sysctl setting on worker nodes must be higher than the operating system default.

Upgrading the operator

If asked, approve the install plan

Upgrading the operand

Upgrading the traced integration capabilities

1️ ⃣ Upgrade operator using Operator Lifecycle Manager.

The Operator Lifecycle Manager (OLM) manages the upgrade of the Operations Dashboard operator in OpenShift.

This ensures that the latest version is available for managing operands.

2️ ⃣ If asked, approve the Install Plan.

Some installations require manual approval of the Install Plan to proceed with the operator upgrade.

If configured for automatic updates, this step may not be required.

3️ ⃣ Upgrade the operand.

Once the operator is upgraded, the operand (Operations Dashboard instance) needs to be updated to the latest version.

This step ensures that the upgraded operator manages the most recent operand version.

4️ ⃣ Upgrade traced integration capabilities.

Finally, upgrade any traced integration capabilities that depend on the Operations Dashboard.

This step ensures compatibility and full functionality with the updated components.

In IBM Cloud Pak for Integration (CP4I) v2021.2 , the Operations Dashboard provides tracing and monitoring for integration capabilities. The correct upgrade sequence ensures a smooth transition with minimal downtime:

Upgrade the Operator using OLM – The Operator manages operands and must be upgraded first.

Approve the Install Plan (if required) – Some operator updates require manual approval before proceeding.

Upgrade the Operand – The actual Operations Dashboard component is upgraded after the operator.

Upgrade Traced Integration Capabilities – Ensures all monitored services are compatible with the new Operations Dashboard version.

IBM Cloud Pak for Integration (CP4I) v2021.2 Administration References:

Upgrading Operators using Operator Lifecycle Manager (OLM)

IBM Cloud Pak for Integration Operations Dashboard

Best Practices for Upgrading CP4I Components

IBM Cloud Pak for Integration (CP4I) includes Confluent Platform as a key capability to support event-driven architecture and real-time data streaming. The Confluent Platform is built on Apache Kafka, providing robust event-streaming capabilities that allow organizations to collect, process, store, and manage data from multiple sources in a highly scalable and reliable manner.

This capability is essential for real-time analytics, event-driven microservices, and data integration between various applications and services. With its high-performance messaging backbone, it ensures low-latency event processing while maintaining fault tolerance and durability.

A. It provides the ability to trace transactions through IBM Cloud Pak for Integration.

Incorrect. Transaction tracing and monitoring are primarily handled by IBM Cloud Pak for Integration ' s API Connect, App Connect, and Instana monitoring tools , rather than Confluent Platform itself.

B. It provides a capability that allows users to store, manage, and retrieve integration assets in IBM Cloud Pak for Integration.

Incorrect. IBM Asset Repository and IBM API Connect are responsible for managing integration assets, not Confluent Platform.

C. It provides APIs to discover applications, platforms, and infrastructure in the environment.

Incorrect. This functionality is more aligned with IBM Instana, IBM Cloud Pak for Multicloud Management, or OpenShift Discovery APIs , rather than the event-streaming capabilities of Confluent Platform.

IBM Cloud Pak for Integration Documentation - Event Streams (Confluent Platform Integration)

IBM Cloud Docs

Confluent Platform Overview

Confluent Documentation

IBM Event Streams for IBM Cloud Pak for Integration

IBM Event Streams

Explanation of Other Options: References:

In IBM Cloud Pak for Integration (CP4I) v2021.2 , the IBM License Service collects and retains license usage data for a default period of 90 days . This data is crucial for auditing and compliance, ensuring that software usage aligns with licensing agreements.

The IBM License Service continuously collects and stores licensing data.

By default, it retains data for 90 days before older data is automatically removed.

Users can query and retrieve usage reports from this 90-day period.

The License Service supports regulatory compliance by ensuring transparent tracking of software usage.

B. The full period from the deployment – Incorrect. The License Service does not retain data indefinitely; it follows a rolling 90-day retention policy.

C. 30 days – Incorrect. The default retention period is longer than 30 days.

D. 60 days – Incorrect. The default is 90 days , not 60.

IBM License Service Documentation

IBM Cloud Pak for Integration v2021.2 – Licensing Guide

IBM Support – License Service Data Retention Policy

Key Details About the IBM License Service Data Retention: Why Not the Other Options? IBM Cloud Pak for Integration (CP4I) v2021.2 Administration References

In IBM Cloud Pak for Integration (CP4I) v2021.2 , ensuring high availability (HA) requires certain components to maintain a quorum . A quorum is a mechanism where a majority of nodes or instances must agree on a state to prevent split-brain scenarios and ensure consistency.

IBM MQ Multi-instance Queue Manager is designed for high availability.

It runs in an active-standby configuration where a shared storage is required, and a quorum ensures that failover occurs correctly .

If the primary queue manager fails, quorum logic ensures that another instance assumes control without data corruption .

API Connect operates in a distributed cluster architecture where multiple components (such as the API Manager, Analytics, and Gateway) work together.

A quorum is required to ensure consistency and avoid conflicts in API configurations across multiple instances.

API Connect uses MongoDB as its backend database , and MongoDB requires a replica set quorum for high availability and failover.

Why " Multi-instance Queue Manager " (A) Requires a Quorum? Why " API Management (API Connect) " (B) Requires a Quorum?

Why Not the Other Options? Option

Reason for Exclusion

C. Application Integration (App Connect)

While App Connect can be deployed in HA mode, it does not require a quorum. It uses Kubernetes scaling and load balancing instead.

D. Event Gateway Service

Event Gateway is stateless and relies on horizontal scaling rather than quorum-based HA.

E. Automation Assets

This component stores automation-related assets but does not require quorum for HA. It typically relies on persistent storage replication.

Thus, Multi-instance Queue Manager (IBM MQ) and API Management (API Connect) require quorum to ensure high availability in Cloud Pak for Integration .

IBM MQ Multi-instance Queue Manager HA

IBM API Connect High Availability and Quorum

CP4I High Availability Architecture

MongoDB Replica Set Quorum in API Connect

IBM Cloud Pak for Integration (CP4I) v2021.2 Administration References:

OpenShift Pipelines , which are based on Tekton , use various components to define and execute CI/CD workflows. The fundamental building block for creating a Pipeline in OpenShift Pipelines is a Task .

Task ( ✅ Correct Answer)

A Task is the basic unit of work in Tekton.

Each Task defines a set of steps (commands) that are executed in containers .

Multiple Tasks are combined into a Pipeline to form a CI/CD workflow.

Pipeline (uses multiple Tasks)

A Pipeline is a collection of Tasks that define the entire CI/CD workflow.

Each Task in the Pipeline runs in sequence or in parallel as specified.

Key Tekton Components:

Why the Other Options Are Incorrect? Option

Explanation

Correct?

A. TaskRun

❌ Incorrect – A TaskRun is an execution instance of a Task , but it does not define the Pipeline itself.

❌

C. TPipe

❌ Incorrect – No such Tekton component called TPipe exists.

❌

D. Pipe

❌ Incorrect – The correct term is Pipeline , not " Pipe " . OpenShift Pipelines do not use this term.

❌

Final Answer: ✅ B. Task

OpenShift Pipelines (Tekton) Documentation

Tekton Documentation – Understanding Tasks

IBM Cloud Pak for Integration – CI/CD with OpenShift Pipelines

IBM Cloud Pak for Integration (CP4I) v2021.2 Administration References: