# EDA Bastion install Gemini-cli and MCP

## Environment

Bastion of RHDPS EDA Bastion
- RHEL 9

## Install Gemini-CLI
1. Install Node.JS

# Download and install nvm:
curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.3/install.sh | bash

# in lieu of restarting the shell
\. "$HOME/.nvm/nvm.sh"
# Download and install Node.js:
nvm install 22

# Verify the Node.js version:
node -v # Should print "v22.17.1".
nvm current # Should print "v22.17.1".

# Verify npm version:
npm -v # Should print "10.9.2".

3. Install Gemini-CLI

npm install -g @google/gemini-cli

4. Get your API-Key
https://aistudio.google.com/app/apikey

export GEMINI_API_KEY=""

5. Edit Gemini setting

mkdir .gemini
vi ~/.gemini/settings.json

{
  "theme": "GitHub",
  "mcpServers": {
    "kubernetes": {
      "command": "npx",
      "args": [
        "-y",
        "rh-tam-kubernetes-mcp-server@latest"
      ]
    }
  }
}

## Test Gemini Prompt mode + MCP

[quickcluster@upi-0 ~]$ vi ~/.gemini/settings.json
[quickcluster@upi-0 ~]$
[quickcluster@upi-0 ~]$
[quickcluster@upi-0 ~]$ gemini -p "列出所有OCP內的namesapce"
OK. Here are all the namespaces in the cluster:
- default
- jace-test
- kube-node-lease
- kube-public
- kube-system
- openshift
- openshift-apiserver
- openshift-apiserver-operator
- openshift-authentication
- openshift-authentication-operator
- openshift-catalogd
- openshift-cloud-controller-manager
- openshift-cloud-controller-manager-operator
- openshift-cloud-credential-operator
- openshift-cloud-network-config-controller
- openshift-cloud-platform-infra
- openshift-cluster-csi-drivers
- openshift-cluster-machine-approver
- openshift-cluster-node-tuning-operator
- openshift-cluster-olm-operator
- openshift-cluster-samples-operator
- openshift-cluster-storage-operator
- openshift-cluster-version
- openshift-config
- openshift-config-managed
...
...

## Ansible playbook呼叫gemini-cli MCP

1. Add Bastion as a new host in AAP inventory


2. Add host in the inventory




3. Create credential for bastion

Rulebook: oc_inspect_analyze.yml

---
- name: Listen for unhealthy+warning event
  hosts: all
  sources:
    - sabre1041.eda.k8s:
        api_version: v1
        kind: Event
        namespace: jace #自行替換成新預計的ns名稱
  rules:
    - name: Debug
      condition: 
        event.resource.reason == "Unhealthy" and event.resource.type == "Warning"
      throttle:
        once_within: 5 minutes
        group_by_attributes:
          - event.resource.metadata.namespace
          - event.resource.involvedObject.name
      actions:
        - run_job_template:
            name: oc-inspect #必須對應AAP內的template 名稱
            organization: Default
        - run_job_template:
            name: gemini-analyze
            organization: Default

Playbook

- name: Run gemini to troubleshoot given NS
  hosts: bastion.xnfx4.sandbox1562.opentlc.com  # 替換成實際的 bastion 主機名或 group
  gather_facts: no
  vars:
    ns: "{{ ansible_eda.event.resource.metadata.namespace }}"

  tasks:
    - name: Gemini analyze NS
      shell: 'gemini -p "分析目前OpenShift內namespace {{ ns }} 有什麼異常 (只需要重點整理，不顯示推論過程)"'
      register: lsout
    - name: Show command output
      debug:
        var: lsout.stdout_lines