Skip to main content
Glama
redhat-community-ai-tools

IPI Provisioner MCP Server

Official

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
MCP_HOSTNoHost for SSE transport (default: 127.0.0.1)127.0.0.1
MCP_PORTNoPort for SSE transport (default: 8000)8000
SSH_KEY_PATHNoPath to SSH public key file (e.g., ~/.ssh/id_rsa.pub)
MCP_TRANSPORTNoMCP transport mode (stdio or sse, default: stdio)stdio
PULL_SECRET_PATHNoPath to pull secret JSON file
OPENSHIFT_RELEASE_IMAGENoOpenShift release image URL (e.g., quay.io/openshift-release-dev/ocp-release:4.17.0-x86_64)

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{
  "listChanged": true
}
prompts
{
  "listChanged": false
}
resources
{
  "subscribe": false,
  "listChanged": false
}
experimental
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
validate_install_configA

Validate an install-config.yaml file.

Args: config_yaml: The install-config.yaml content as a string

Returns: Validation results with errors and summary

generate_install_configA

Generate an install-config.yaml for baremetal IPI.

Args: cluster_name: Name of the cluster base_domain: Base domain (e.g., example.com) pull_secret: Pull secret content (JSON string) ssh_key: SSH public key content api_vip: API VIP address ingress_vip: Ingress VIP address hosts: List of baremetal hosts with BMC configuration machine_network_cidr: Machine network CIDR (default: 192.168.1.0/24) cluster_network_cidr: Cluster network CIDR (default: 10.128.0.0/14) service_network_cidr: Service network CIDR (default: 172.30.0.0/16) network_type: Network type (default: OVNKubernetes) control_plane_replicas: Number of master nodes (default: 3) worker_replicas: Number of worker nodes (default: 0)

Returns: Generated install-config as dictionary and YAML string

get_config_templateA

Get an install-config template.

Args: template_name: Template name (e.g., 'baremetal-ipi-compact', 'baremetal-ipi-ha')

Returns: Template content and metadata

list_config_templatesA

List all available install-config templates.

Returns: List of templates with descriptions

create_baremetal_host_entryA

Create a baremetal host entry for install-config.yaml.

Args: name: Host name (e.g., 'master-0') role: Host role ('master' or 'worker') bmc_address: BMC address (e.g., 'redfish://192.168.1.100') bmc_username: BMC username bmc_password: BMC password boot_mac_address: Boot MAC address (e.g., 'aa:bb:cc:dd:ee:ff') root_device_hint: Optional root device hint (e.g., '/dev/sda' or 'wwn')

Returns: Host configuration dictionary ready for install-config.yaml

test_bmc_connectionA

Test BMC connectivity (Redfish).

Args: bmc_address: BMC address (e.g., 'redfish://192.168.1.100' or 'redfish+https://192.168.1.100') username: BMC username password: BMC password verify_ssl: Whether to verify SSL certificates (default: False)

Returns: Connection test results

validate_bmc_addressesA

Validate BMC addresses for multiple hosts.

Args: hosts: List of host configurations with BMC details Each host should have: name, bmc.address, bmc.username, bmc.password

Returns: Validation results for all hosts

parse_bmc_addressA

Parse and validate BMC address format.

Args: bmc_address: BMC address (e.g., 'redfish://192.168.1.100:443')

Returns: Parsed BMC address components

generate_bmc_test_scriptB

Generate a bash script to test BMC connectivity for multiple hosts.

Args: hosts: List of host configurations with BMC details

Returns: Bash script for testing BMC connectivity

validate_network_cidrA

Validate CIDR notation.

Args: cidr: CIDR string (e.g., '192.168.1.0/24')

Returns: Validation result

check_ip_in_networkB

Check if an IP address is within a CIDR range.

Args: ip_address: IP address to check cidr: CIDR range

Returns: Check result

check_network_overlapB

Check if two networks overlap.

Args: cidr1: First CIDR cidr2: Second CIDR

Returns: Overlap check result

validate_vip_configurationC

Validate VIP (Virtual IP) configuration.

Args: api_vip: API VIP address ingress_vip: Ingress VIP address machine_network_cidr: Machine network CIDR

Returns: Validation results

generate_dns_configA

Generate DNS configuration for the cluster.

Args: cluster_name: Cluster name base_domain: Base domain api_vip: API VIP address ingress_vip: Ingress VIP address hosts: Optional JSON string of hosts list. Each host should have 'name' and 'ip' keys. Example: '[{"name": "master-0", "ip": "10.1.235.25"}, {"name": "master-1", "ip": "10.1.235.26"}]'

Returns: Required DNS records and validation commands including A and PTR records for hosts

calculate_cluster_network_capacityA

Calculate cluster network capacity.

Args: cluster_cidr: Cluster network CIDR (e.g., '10.128.0.0/14') host_prefix: Host prefix - subnet size per node (e.g., 23)

Returns: Capacity calculations (max nodes, IPs per node, etc.)

generate_network_diagramA

Generate a network diagram for the cluster.

Args: cluster_cidr: Cluster (pod) network CIDR service_cidr: Service network CIDR machine_cidr: Machine network CIDR api_vip: API VIP ingress_vip: Ingress VIP

Returns: Network diagram

validate_complete_network_configA

Validate complete network configuration for overlaps and validity.

Args: cluster_cidr: Cluster network CIDR service_cidr: Service network CIDR machine_cidr: Machine network CIDR api_vip: Optional API VIP (required for HA) ingress_vip: Optional Ingress VIP (required for HA)

Returns: Comprehensive validation results

check_installer_binaryA

Check if openshift-install binary is available.

Args: install_dir: Optional installation directory to check for local binary

Returns: Binary existence check results

extract_openshift_installerA

Extract openshift-install binary from a release image.

Args: release_image: OpenShift release image URL (e.g., 'quay.io/openshift-release-dev/ocp-release:4.17.0-x86_64') output_dir: Directory to extract installer to (default: current directory)

Returns: Extraction results and next steps

create_installation_manifestsA

Create Kubernetes manifests from install-config.yaml.

Args: install_dir: Installation directory containing install-config.yaml installer_path: Path to openshift-install binary (default: 'openshift-install')

Returns: Manifest creation results

prepare_cluster_installationA

Prepare cluster installation command.

This tool doesn't actually start the installation (which can take 30-60 minutes), but provides the command to run.

Args: install_dir: Installation directory installer_path: Path to openshift-install binary (default: 'openshift-install') log_level: Log level (debug, info, warn, error)

Returns: Installation command and information

prepare_cluster_destroyA

Prepare cluster destroy command.

This tool provides the command to destroy the cluster. Use with caution - this will delete the entire cluster!

Args: install_dir: Installation directory installer_path: Path to openshift-install binary (default: 'openshift-install')

Returns: Destroy command and information

get_installation_statusB

Get cluster installation status.

Args: install_dir: Installation directory

Returns: Installation status and progress

get_installation_logsA

Get recent installation log entries.

Args: install_dir: Installation directory lines: Number of recent lines to retrieve (default: 50)

Returns: Recent log entries

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/redhat-community-ai-tools/openshift-ipi-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server