Provision OpenShift on IBM DevIT FYRE¶

Overview¶

The MAS CLI provides automated provisioning of OpenShift Container Platform clusters on IBM's internal FYRE infrastructure. FYRE offers rapid cluster deployment for development, testing, and demonstration purposes with flexible resource allocation and quota management.

This guide covers the complete process of provisioning an OpenShift cluster on FYRE, from obtaining credentials to configuring storage providers.

Preparation¶

FYRE Account and Credentials¶

You must have an active FYRE account with appropriate quota allocation. To obtain FYRE credentials:

1. Access the FYRE Portal
2. Log in with your IBM credentials
3. Navigate to Account Settings to retrieve your API key
4. Note your username and API key for use with the CLI

Product Group and Quota¶

FYRE uses product groups to organize and allocate resources. You will need:

• Product Group ID - The numeric identifier for your product group
• Quota Type - Either quick_burn (temporary, time-limited) or product_group (allocated quota)

Contact your FYRE administrator if you need assistance with product group access or quota allocation.

OpenShift Version Selection¶

Choose an OpenShift version supported by both FYRE and your target MAS version. Refer to the MAS system requirements for version compatibility.

Cluster Configuration¶

Cluster Naming¶

Cluster names must:

• Use lowercase letters only
• Be unique within your FYRE account
• Be descriptive for easy identification
• Follow your organization's naming conventions

Example: mas-dev-cluster, test-env-01

Worker Node Sizing¶

Configure worker nodes based on your workload requirements:

Configuration	CPU	Memory	Use Case
Small	4	16 GB	Development, testing
Medium	8	32 GB	Standard MAS installation
Large	16	64 GB	Production-like environments

Additional Storage Disks¶

FYRE allows attaching additional disks to worker nodes for storage providers like ODF or Longhorn. Specify disk sizes as a comma-separated list:

• 200,200 - Two 200 GB disks per worker
• 500 - One 500 GB disk per worker

Additional disks are required when using ODF (OpenShift Data Foundation) for persistent storage.

Storage Provider Options¶

The CLI can automatically configure storage providers:

• NFS - Network File System, simple and fast setup
• ODF - OpenShift Data Foundation, production-grade storage
• Longhorn - Cloud-native distributed storage

Provisioning Modes¶

Interactive Mode¶

Interactive mode guides you through the provisioning process with prompts for all configuration options.

docker run -ti --rm --pull always quay.io/ibmmas/cli mas provision-fyre

The interactive session will:

1. Prompt for FYRE credentials
2. Request cluster configuration details
3. Configure worker node specifications
4. Optionally configure storage provider
5. Display a summary and request confirmation

Non-Interactive Mode¶

Non-interactive mode is ideal for automation and scripting. All required parameters must be provided via command-line arguments or environment variables.

export FYRE_USERNAME=your-username
export FYRE_APIKEY=your-api-key

docker run -ti --rm --pull always quay.io/ibmmas/cli mas provision-fyre \
  -u $FYRE_USERNAME \
  -a $FYRE_APIKEY \
  -p 225 \
  -q product_group \
  -c mas-dev-cluster \
  -d "MAS Development Cluster" \
  -v 4.15 \
  --worker-count 3 \
  --worker-cpu 8 \
  --worker-memory 32 \
  --no-confirm

Configuration Examples¶

Basic Development Cluster¶

Minimal configuration for development and testing:

docker run -ti --rm --pull always quay.io/ibmmas/cli mas provision-fyre \
  -u $FYRE_USERNAME \
  -a $FYRE_APIKEY \
  -p 225 \
  -q quick_burn \
  -c dev-cluster \
  -v 4.15 \
  --worker-count 3 \
  --worker-cpu 4 \
  --worker-memory 16 \
  --storage nfs \
  --no-confirm

Standard MAS Installation Cluster¶

Recommended configuration for typical MAS installations:

docker run -ti --rm --pull always quay.io/ibmmas/cli mas provision-fyre \
  -u $FYRE_USERNAME \
  -a $FYRE_APIKEY \
  -p 225 \
  -q product_group \
  -c mas-standard \
  -d "MAS Standard Installation" \
  -v 4.15 \
  --worker-count 3 \
  --worker-cpu 8 \
  --worker-memory 32 \
  --storage nfs \
  --nfs-image-registry-size 100 \
  --no-confirm

Cluster with ODF Storage¶

Configuration with OpenShift Data Foundation for production-like storage:

docker run -ti --rm --pull always quay.io/ibmmas/cli mas provision-fyre \
  -u $FYRE_USERNAME \
  -a $FYRE_APIKEY \
  -p 225 \
  -q product_group \
  -c mas-odf-cluster \
  -d "MAS Cluster with ODF" \
  -v 4.15 \
  --worker-count 3 \
  --worker-cpu 16 \
  --worker-memory 64 \
  --worker-additional-disks "200,200" \
  --storage odf \
  --no-confirm

Large Cluster for Performance Testing¶

High-resource configuration for performance testing:

docker run -ti --rm --pull always quay.io/ibmmas/cli mas provision-fyre \
  -u $FYRE_USERNAME \
  -a $FYRE_APIKEY \
  -p 225 \
  -q product_group \
  -c mas-perf-test \
  -d "MAS Performance Testing" \
  -v 4.15 \
  --worker-count 5 \
  --worker-cpu 16 \
  --worker-memory 64 \
  --worker-additional-disks "500,500" \
  --storage odf \
  --no-confirm

Command Reference¶

FYRE Credentials¶

• -u, --username FYRE_USERNAME - FYRE username (required)
• -a, --apikey FYRE_APIKEY - FYRE API key (required)

Cluster Configuration¶

• -p, --product-id FYRE_PRODUCT_ID - FYRE product group ID (required)
• -q, --quota-type FYRE_QUOTA_TYPE - Quota type: quick_burn or product_group (required)
• -c, --cluster-name CLUSTER_NAME - Cluster name, lowercase only (required)
• -v, --ocp-version OCP_VERSION - OpenShift version, e.g., 4.14, 4.15 (required)
• -d, --description FYRE_DESCRIPTION - Cluster description (optional)

Worker Node Configuration¶

• --worker-count FYRE_WORKER_COUNT - Number of worker nodes (default: 3)
• --worker-cpu FYRE_WORKER_CPU - CPUs per worker node (default: 8)
• --worker-memory FYRE_WORKER_MEMORY - Memory per worker in GB (default: 32)
• --worker-additional-disks FYRE_WORKER_ADDITIONAL_DISKS - Comma-separated disk sizes in GB (optional)
• --fyre-cluster-size FYRE_CLUSTER_SIZE - Quick burn size: medium or large (optional)

Storage Configuration¶

• --storage - Storage provider: nfs, odf, or longhorn (optional)
• --nfs-image-registry-size FYRE_NFS_IMAGE_REGISTRY_SIZE - NFS registry size in GB (default: 100)

Other Options¶

• --no-confirm - Skip confirmation prompt
• -h, --help - Display help message

Post-Provisioning Steps¶

Accessing Your Cluster¶

After provisioning completes, the CLI will display:

1. Cluster URL - OpenShift console access URL
2. Kubeadmin Credentials - Initial admin username and password
3. API Server URL - For CLI access

Save these credentials securely.

Connecting with oc CLI¶

oc login --server=https://api.your-cluster.fyre.ibm.com:6443 \
  --username=kubeadmin \
  --password=<provided-password>

Verifying Cluster Health¶

Check that all nodes are ready:

oc get nodes

Verify storage classes are available:

oc get storageclass

Next Steps¶

With your FYRE cluster provisioned, you can proceed to:

• Install MAS
• Configure additional cluster resources
• Set up monitoring and logging

Troubleshooting¶

Provisioning Failures¶

Quota Exceeded

Error: Insufficient quota available

• Verify your product group has available quota
• Contact your FYRE administrator to increase quota
• Use quick_burn quota for temporary clusters

Invalid Cluster Name

Error: Cluster name must be lowercase

• Ensure cluster name contains only lowercase letters, numbers, and hyphens
• Avoid uppercase letters and special characters

Version Not Available

Error: OpenShift version not supported

• Check available versions in the FYRE portal
• Use a supported version (typically 4.14, 4.15, or 4.16)

Storage Configuration Issues¶

NFS Registry Size Too Large

Error: Requested size exceeds available storage

• Reduce --nfs-image-registry-size value
• Default 100 GB is usually sufficient

ODF Requires Additional Disks

Error: ODF requires additional disks on worker nodes

• Add --worker-additional-disks parameter
• Minimum two disks recommended: --worker-additional-disks "200,200"

Connection Issues¶

Unable to Access Cluster - Verify you are on IBM network or VPN - Check cluster status in FYRE portal - Ensure firewall rules allow access

Authentication Failures - Verify credentials are correct - Check if password has expired - Try regenerating kubeadmin password in FYRE portal

Best Practices¶

Resource Planning¶

• Development: 3 workers, 4 CPU, 16 GB memory
• Testing: 3 workers, 8 CPU, 32 GB memory
• Production-like: 5+ workers, 16 CPU, 64 GB memory

Quota Management¶

• Use quick_burn for short-term testing (auto-expires)
• Use product_group quota for longer-term environments
• Monitor quota usage regularly
• Clean up unused clusters promptly

Naming Conventions¶

• Include purpose in name: mas-dev, mas-test, mas-demo
• Add date for temporary clusters: mas-test-20260520
• Use team or project identifiers: team-a-mas-dev

Storage Selection¶

• NFS: Fast setup, good for development
• ODF: Production-grade, requires more resources
• Longhorn: Alternative to ODF, lighter weight