devops/containers/podman/SKILL.md
Manage containers using Podman, the daemonless container engine. Run rootless containers, create pods, manage images, and use Docker-compatible commands. Use when working with Podman or requiring rootless container operations.
npx skillsauth add bagelhole/devops-security-agent-skills podmanInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Run and manage containers without a daemon using Podman's rootless container engine.
Use this skill when:
| Feature | Docker | Podman | |---------|--------|--------| | Architecture | Client-daemon | Daemonless | | Root required | Default | Optional (rootless) | | Pod support | No | Yes (Kubernetes-style) | | Systemd integration | Limited | Native | | Socket | docker.sock | podman.sock (optional) |
# Run container (identical to Docker)
podman run -d --name webserver -p 8080:80 nginx
# List containers
podman ps -a
# Stop and remove
podman stop webserver
podman rm webserver
# Execute command
podman exec -it webserver /bin/sh
# View logs
podman logs -f webserver
# Pull image
podman pull docker.io/library/nginx:latest
# List images
podman images
# Build image
podman build -t myapp:latest .
# Push to registry
podman push myapp:latest registry.example.com/myapp:latest
# Remove image
podman rmi nginx:latest
# Check user namespace support
cat /proc/sys/user/max_user_namespaces
# Enable if needed (as root)
echo "user.max_user_namespaces=28633" | sudo tee /etc/sysctl.d/userns.conf
sudo sysctl -p /etc/sysctl.d/userns.conf
# Configure subuid/subgid for user
sudo usermod --add-subuids 100000-165535 --add-subgids 100000-165535 $USER
# Verify
podman unshare cat /proc/self/uid_map
# Run as regular user (no sudo)
podman run -d --name myapp -p 8080:80 nginx
# Check user namespace mapping
podman unshare id
# Verify non-root
podman top myapp user
# Rootless cannot bind to ports < 1024 by default
# Use ports >= 1024
podman run -d -p 8080:80 nginx
# Or enable unprivileged ports (as root)
echo "net.ipv4.ip_unprivileged_port_start=80" | sudo tee /etc/sysctl.d/ports.conf
sudo sysctl -p /etc/sysctl.d/ports.conf
# Create pod
podman pod create --name mypod -p 8080:80 -p 5432:5432
# Add containers to pod
podman run -d --pod mypod --name web nginx
podman run -d --pod mypod --name db postgres:15
# List pods
podman pod ps
# Containers share network namespace
podman exec web curl localhost:5432
# Start/stop pod (affects all containers)
podman pod start mypod
podman pod stop mypod
# Remove pod and containers
podman pod rm -f mypod
# View pod details
podman pod inspect mypod
# Generate Kubernetes YAML from pod
podman generate kube mypod > mypod.yaml
# Generate unit file for container
podman generate systemd --new --name myapp > ~/.config/systemd/user/container-myapp.service
# For pod
podman generate systemd --new --name mypod --files
# Reload systemd
systemctl --user daemon-reload
# Enable and start
systemctl --user enable --now container-myapp.service
# ~/.config/containers/systemd/webapp.container
[Container]
Image=docker.io/library/nginx:latest
PublishPort=8080:80
Volume=webapp-data:/usr/share/nginx/html
[Service]
Restart=always
[Install]
WantedBy=default.target
# Reload to generate service
systemctl --user daemon-reload
# Start the service
systemctl --user start webapp
# Install podman-compose
pip install podman-compose
# Run compose file
podman-compose up -d
# Or use Docker Compose with Podman socket
systemctl --user enable --now podman.socket
export DOCKER_HOST=unix:///run/user/$UID/podman/podman.sock
docker-compose up -d
# Play Kubernetes YAML
podman kube play deployment.yaml
# Stop and remove
podman kube down deployment.yaml
# Create network
podman network create mynetwork
# Run on network
podman run -d --network mynetwork --name app myapp
# Connect container to network
podman network connect mynetwork existing-container
# List networks
podman network ls
# Inspect network
podman network inspect mynetwork
# Containers on same network can resolve by name
podman run -d --network mynetwork --name db postgres:15
podman run -d --network mynetwork --name app \
-e DATABASE_HOST=db myapp
# Create volume
podman volume create mydata
# Use volume
podman run -d -v mydata:/data myapp
# List volumes
podman volume ls
# Inspect volume
podman volume inspect mydata
# Rootless volumes location
ls ~/.local/share/containers/storage/volumes/
# Bind mount with SELinux label
podman run -v ./data:/app/data:Z myapp
# Z = private label (single container)
# z = shared label (multiple containers)
# Edit registries.conf
# ~/.config/containers/registries.conf
unqualified-search-registries = ["docker.io", "quay.io"]
[[registry]]
prefix = "docker.io"
location = "docker.io"
[[registry.mirror]]
location = "mirror.gcr.io"
# Login to registry
podman login docker.io
# Login to private registry
podman login registry.example.com
# Credentials stored in
# ~/.config/containers/auth.json
# Podman uses Buildah for builds
podman build -t myapp:latest .
# Build with specific format
podman build --format docker -t myapp .
# Multi-stage build
podman build --target production -t myapp:prod .
# Create container from scratch
buildah from scratch
buildah copy working-container ./app /app
buildah config --entrypoint '["/app/main"]' working-container
buildah commit working-container myapp:minimal
Problem: Cannot access files in mounted volumes
Solution: Use :Z or :z suffix for SELinux, or check ownership
Problem: Port not accessible in rootless mode Solution: Use ports >= 1024 or configure unprivileged port start
Problem: Images download slowly Solution: Configure registry mirrors in registries.conf
Problem: Container doesn't start via systemd
Solution: Enable lingering: loginctl enable-linger $USER
docker to podman for compatibilitydevelopment
Design and operationalize SRE dashboards that surface reliability, latency, error, saturation, and capacity signals across services. Use when building observability views for SLOs, incident response, and executive reliability reporting.
testing
Harden OpenClaw self-hosted environments with baseline host controls, auth tightening, secret handling, network segmentation, and safe update/rollback workflows. Use when deploying OpenClaw in home labs, startups, or production-like local AI infrastructure.
devops
Deploy, manage, and optimize vector databases for AI applications. Covers Qdrant, Weaviate, pgvector, and Pinecone — collection management, indexing strategies, backup, and performance tuning for production RAG and semantic search workloads.
testing
Deploy ML models on Kubernetes with KServe (formerly KFServing) and NVIDIA Triton Inference Server. Includes canary deployments, autoscaling, model versioning, A/B testing, and GPU resource management for production model serving.