Skip to content

OADP.md

Latest commit

 

History

History
105 lines (70 loc) · 3.16 KB

OADP.md

File metadata and controls

105 lines (70 loc) · 3.16 KB
Raw

OADP Support

The Kubernetes MCP Server supports OpenShift API for Data Protection (OADP) resources using the core toolset's generic resource tools. OADP resources (Velero backups, restores, schedules, and related CRDs) can be managed through the standard resources_list, resources_get, resources_create, and resources_delete tools.

The OADP toolset provides a troubleshooting prompt for diagnosing backup and restore issues.

Prompt

oadp-troubleshoot

Generate a step-by-step troubleshooting guide for diagnosing OADP backup and restore issues. Gathers DPA status, BSL health, recent backup/restore status, Velero pod health, pod logs, and events into a single diagnostic workflow.

Arguments:

  • namespace (optional): The OADP namespace (default: openshift-adp)
  • backup (optional): Name of a specific backup to troubleshoot
  • restore (optional): Name of a specific restore to troubleshoot

Example Usage:

Use the oadp-troubleshoot prompt with namespace=openshift-adp and backup=my-failing-backup

Enable the OADP Toolset

Option 1: Command Line

kubernetes-mcp-server --toolsets core,config,oadp

Option 2: Configuration File

toolsets = ["core", "config", "oadp"]

Option 3: MCP Client Configuration

{
  "mcpServers": {
    "kubernetes": {
      "command": "npx",
      "args": ["-y", "kubernetes-mcp-server@latest", "--toolsets", "core,config,oadp"]
    }
  }
}

Prerequisites

OADP support requires:

  1. OADP Operator installed - The OADP operator must be installed on the cluster
  2. DataProtectionApplication configured - At least one DPA must be configured
  3. Proper RBAC - The user/service account must have permissions to access Velero CRDs

Verify OADP Installation

# Check OADP operator
oc get csv -n openshift-adp | grep oadp

# Check DPA
oc get dpa -n openshift-adp

# Check BSL availability
oc get backupstoragelocations -n openshift-adp

Common Use Cases

OADP resources are managed using the core toolset's generic resource tools. Here are common workflows:

Check OADP Health

Use resources_list to check DPA and BSL status:

  • List DPAs: apiVersion: oadp.openshift.io/v1alpha1, kind: DataProtectionApplication
  • List BSLs: apiVersion: velero.io/v1, kind: BackupStorageLocation

Create and Monitor a Backup

  1. Create a backup using resources_create with apiVersion: velero.io/v1, kind: Backup
  2. Monitor status using resources_get

Restore from Backup

  1. List available backups using resources_list
  2. Create a restore using resources_create with apiVersion: velero.io/v1, kind: Restore
  3. Monitor status using resources_get

Set Up Scheduled Backups

Create a schedule using resources_create with apiVersion: velero.io/v1, kind: Schedule

Troubleshooting

Use the oadp-troubleshoot prompt to automatically gather diagnostic information including DPA status, BSL health, Velero pod logs, and recent events.

Common Issues

  • BSL "Unavailable": Check cloud credentials and bucket accessibility
  • Backup stuck "InProgress": Check Velero pod logs for errors
  • DPA not reconciled: Verify the OADP operator pod is running