Skip to content

devops Pod

Standard OCI container - works with Docker, Podman, Kubernetes, Apptainer.

The devops pod provides a comprehensive suite of cloud infrastructure and Kubernetes tools for DevOps workflows, including AWS, Google Cloud, and Grafana observability tools.

Overview

Attribute Value
Image ghcr.io/atrawog/bazzite-ai-pod-devops:stable
Size ~4GB
GPU None (CPU-only)
Inherits pod-base

Quick Start

# Basic usage
docker run -it --rm -v $(pwd):/workspace \
  ghcr.io/atrawog/bazzite-ai-pod-devops:stable

# With AWS and Kubernetes credentials
docker run -it --rm \
  -v $(pwd):/workspace \
  -v ~/.aws:/home/jovian/.aws:ro \
  -v ~/.kube:/home/jovian/.kube:ro \
  ghcr.io/atrawog/bazzite-ai-pod-devops:stable

apiVersion: v1
kind: Pod
metadata:
  name: devops-shell
spec:
  containers:
  - name: devops
    image: ghcr.io/atrawog/bazzite-ai-pod-devops:stable
    command: ["sleep", "infinity"]
    volumeMounts:
    - name: kubeconfig
      mountPath: /home/jovian/.kube
      readOnly: true
  volumes:
  - name: kubeconfig
    secret:
      secretName: kubeconfig

apptainer pull docker://ghcr.io/atrawog/bazzite-ai-pod-devops:stable
apptainer exec bazzite-ai-pod-devops_stable.sif bash

apptainer pull docker://ghcr.io/atrawog/bazzite-ai-pod-devops:stable
apptainer shell bazzite-ai-pod-devops_stable.sif

What's Included

Cloud Providers

Tool Description
AWS CLI v2 Amazon Web Services command-line interface
gcloud Google Cloud SDK
Firebase CLI Firebase development tools
Gemini CLI Google AI integration
Scaleway CLI Scaleway cloud platform
Wrangler Cloudflare Workers CLI

Kubernetes Tools

Tool Description
kubectl Kubernetes cluster management
Helm Kubernetes package manager
kubectx/kubens Context and namespace switchers
OpenTofu Infrastructure as Code (Terraform fork)

Grafana Observability

Tool Description
mcp-grafana Grafana MCP server for Claude Code
logcli Loki log query CLI
promtool Prometheus configuration utility
mimirtool Mimir metrics management
tempo-cli Tempo trace query CLI

From base Pod

  • Python, Node.js, Go, Rust
  • VS Code, Docker CLI, Podman
  • Build tools, Claude Code

Usage

AWS Configuration

# Inside the pod
aws configure
# Enter: Access Key ID, Secret Access Key, Region, Output format

# Verify
aws sts get-caller-identity
aws s3 ls

Google Cloud

# Authenticate
gcloud auth login

# Set project
gcloud config set project my-project-id

# List resources
gcloud compute instances list
gcloud container clusters list

Kubernetes

# Set context (if kubeconfig mounted)
kubectl config get-contexts
kubectx my-cluster

# Common operations
kubectl get pods --all-namespaces
kubectl apply -f deployment.yaml
kubectl logs -f deployment/my-app

Helm Charts

# Add a repository
helm repo add bitnami https://charts.bitnami.com/bitnami
helm repo update

# Install a chart
helm install my-release bitnami/nginx

# List releases
helm list

OpenTofu (Infrastructure as Code)

# Initialize
cd /workspace/terraform
tofu init

# Plan changes
tofu plan -out=plan.tfplan

# Apply
tofu apply plan.tfplan

# Destroy
tofu destroy

Scaleway

# Initialize
scw init

# List resources
scw instance server list
scw k8s cluster list

Credentials Management

AWS Credentials

Mount your AWS credentials:

docker run -it --rm \
  -v ~/.aws:/home/jovian/.aws:ro \
  -v $(pwd):/workspace \
  ghcr.io/atrawog/bazzite-ai-pod-devops:stable

Kubeconfig

Mount your kubeconfig:

docker run -it --rm \
  -v ~/.kube:/home/jovian/.kube:ro \
  -v $(pwd):/workspace \
  ghcr.io/atrawog/bazzite-ai-pod-devops:stable

Environment Variables

# Pass secrets via environment
docker run -it --rm \
  -e AWS_ACCESS_KEY_ID=xxx \
  -e AWS_SECRET_ACCESS_KEY=xxx \
  -e AWS_DEFAULT_REGION=us-east-1 \
  -v $(pwd):/workspace \
  ghcr.io/atrawog/bazzite-ai-pod-devops:stable

Common Workflows

Deploy to Kubernetes

# 1. Build and push image
docker build -t my-registry/my-app:v1 .
docker push my-registry/my-app:v1

# 2. Deploy with Helm
helm upgrade --install my-app ./chart \
  --set image.tag=v1 \
  --namespace production

# 3. Verify
kubectl rollout status deployment/my-app -n production
kubectl get pods -n production

Infrastructure Changes

# 1. Format and validate
tofu fmt
tofu validate

# 2. Plan
tofu plan -out=plan.tfplan

# 3. Review plan output
# 4. Apply
tofu apply plan.tfplan

# 5. Commit state changes

Troubleshooting

kubectl: connection refused

Ensure kubeconfig is mounted and valid:

# Check kubeconfig location
echo $KUBECONFIG

# Verify cluster access
kubectl cluster-info

# If using mounted config
kubectl --kubeconfig=/home/jovian/.kube/config get nodes

AWS: Invalid credentials

# Verify credentials
aws sts get-caller-identity

# Check environment variables
env | grep AWS

# Reconfigure
aws configure

gcloud: Not authenticated

# Re-authenticate
gcloud auth login
gcloud auth application-default login

# Verify
gcloud auth list

See Also